The present invention concerns a method and a device for encoding image data. The invention further relates to a method and a device for decoding image data, and a bitstream of video data.

Video data is typically composed of a series of still images which are shown rapidly in succession as a video sequence to give the idea of a moving image. Video applications are continuously moving towards improved image resolution (greater number of pixels per frame, higher frame rate, higher bit-depth ...). A large quantity of video content is distributed in digital form via broadcast channels, digital networks and packaged media, with a continuous evolution towards improved quality and resolution (e.g. higher number of pixels per frame, higher frame rate, higher bit-depth or extended colour gamut). This evolution in technology puts increased pressure on distribution networks that already face difficulties in providing HDTV resolution and data rates economically to the end user. Consequently, further increases in data rate will put additional pressure on such distribution networks. To address this challenge, ITU-T (International Telecommunications Union, telecommunications Standardization Sector) and ISO/MPEG decided to launch a new video coding standard project in January 2010, known as High Efficiency Video Coding (HEVC). Note that in the following the term "HEVC" Is used to represent the current implementation of HEVC, which is in course of standardization and will be subject of evolutions.

HEVC codec design is similar to that of most previous so-called block-based hybrid transform codecs such as H.263, H.264, MPEG-1 , MPEG-2, MPEG-4, SVC. Video compression algorithms, such as those standardized by the standardization bodies ITU, ISO and SMPTE, use spatial and temporal redundancies of images in order to generate data bit streams of reduced size. Spatial redundancy represents the mutual correlation between adjacent image pixels, while temporal redundancy represents the correlation between images of sequential images. Such compression processes make the transmission and/or storage of video sequences more effective.

Video sequences encoded by HEVC will be used in many applications such as streaming environments or lightweight performance devices. I n order to be more robust to streaming errors and to be able to process data in parallel in lightweight devices, for example, new coding structures and rules have been added to HEVC. However, current HEVC techniques do not provide a solution that is flexible enough to respond to requirements of all of these applications efficiently.

The present invention has been devised to address one or more of the foregoing concerns. According to a first aspect of the invention there is provided a method of encoding an image of a video sequence, the image being represented by a plurality of coding units, the encoding method comprising at least two different encoding processes for encoding the image, wherein each encoding process introduces coding dependencies between encoded data associated with the respective encoding process, the method comprising: determining, for each encoding process, a set of slice boundaries wherein the respective set of slice boundaries partitions the image into one or more slices for the respective encoding process, the one or more slices each including a set of coding units of the image between a pair of respective slice boundaries, wherein the coding units of one slice are encoded independently of the coding units of another of the slices of the encoding process concerned.

Accordingly more flexible video processing is provided since a decoder receiving the encoded image may perform processing on the different types of slices according to its decoding capabilities. The slice boundaries may be adapted to the application or the characteristics of the processing modules. By encoding the image such that coding dependencies between coding units of one slice and coding units of another of the slices of the same encoding process are prevented, decoding processes of the same type may be performed in parallel independently. The slices defined by the slice boundaries may be adapted according to the corresponding decoding process to be performed.

In one embodiment, the coding units of one slice are encoded independently of the coding units of another of the slices of the encoding process concerned, for at least one set of slices. In another embodiment, for each set of slices the coding units of one slice are encoded independently of the coding units of another of the slices of the encoding process concerned.

In a particular embodiment of the invention a slice information dataset may be provided for the image including slice information defining each type of encoding process and the corresponding slice boundaries within the image for each type of encoding process. Accordingly, the slice information does not need to be duplicated in slice data thereby reducing the number of bits needed to send the video stream and thus the data overhead. Moreover a decoder may decode the received data without having to decode the slice data.

In an embodiment the slice information dataset may be transmitted prior to the slices (encoded image). Thus the decoder can prepare for the decoding processes in accordance with the decoding information provided in the slice information dataset and allocate its resources accordingly. For example the decoder may adapt the number of processes that can be performed in parallel.

In an embodiment the slice information dataset is transmitted in a separate network abstraction unit to the network abstraction units transmitting the slices. This enables the slice information dataset to be protected in a more robust manner independently of the slices. The slice information dataset may be included in a sequence parameter set of the video sequence or a picture parameter set of the image.

The slices for each respective encoding process may be defined by the slice boundaries to be equal in size.

In a particular embodiment at least one set of slice boundaries defines one or more slices equal in size to the maximum size of a transmission unit of the medium in which the video sequence is to be transmitted. This enables the packetization process to be simplified.

The slice information dataset may also include parameter data defining the respective coding parameters of each encoding process. In a particular embodiment one of the encoding processes comprises at least one of an entropy coding process and motion information prediction process. The slice information dataset may thus include information indicating if the encoding process comprises context-adaptive binary arithmetic coding or context-adaptive variable-length coding.

Another of the different encoding processes may comprise intra prediction coding for pixel sample reconstruction. The slice information dataset may then include information indicating the quantization parameter used to quantize the residual of the video sequence.

A further encoding process may comprise a loop filtering process. The slice information dataset may then include information indicating the location of a loop filtering data slice where loop filtering is disabled The slice information dataset may include adaptive loop filter data defining adaptive loop filter coefficients. In a particular embodiment the slice boundaries are defined according to at least one of the video content, capability of the decoder to which the video sequence is to be transmitted and characteristics of the network via which the video sequence is to be transmitted. This enables an adaptable flexible video streaming system to be provided.

The slice information data set may be protected for transmission by an error correction technique. This ensures that the slice information data is transmitted more robustly and the probability of slice information loss is reduced;

In a further embodiment of the invention one or more individual parameters subsets are provided for each type of encoding process. This enables parameters for each different encoding process to be gathered in a single NAL unit for example. Each individual parameter subset may be transmitted in a separate network abstraction unit to the network abstraction units transmitting the slices.

According to a second aspect of the invention there is provided a method of decoding a video bitstream representative of an image of a video sequence, the image being represented by a plurality of coding units and having been encoded by at least two different encoding processes, wherein each encoding process introduces coding dependencies between encoded data associated with the respective encoding process, the method comprising decoding the bitstream according to sets of slice boundaries wherein each set of slice boundaries corresponds to a different decoding process and defines one or more slices encoded according to one of the different encoding processes, each slice including a set of coding units, wherein the coding units of one slice are decoded independently of the coding units of another of the slices for the same decoding process. Slice boundaries of one type of encoding (or decoding) process may be defined independently of slice boundaries of another type of encoding (or decoding) process. In an embodiment of the invention the decoding method may further include receiving a slice information dataset for each image including data identifying at least two different decoding processes for decoding the image and indicating the respective set of slice boundaries within the image for each decoding process. The slice information dataset may be received prior to reception of the encoded image.

In an embodiment each of the slices of a set of slices of a respective decoding process are decoded in parallel. In another embodiment the different decoding processes are performed in parallel. More efficient decoding may thus be provided.

According to a third aspect of the invention there is provided a bitstream of video data comprising: an encoded image, encoded using at least two different encoding processes wherein each encoding process introduces coding dependencies between encoded data associated with the respective encoding process and is associated with a respective set of slice boundaries for the respective encoding process, wherein the respective set of slice boundaries for each encoding process independently partitions the encoded image into one or more slices, the one or more slices each including a set of coding units of the image between a pair of respective slice boundaries and wherein the coding units of one slice are encoded independently of the coding units of another of the slices of the same encoding process. In a particular embodiment a slice information dataset for the image including slice information defining each type of encoding process and the corresponding slice boundaries within the image for each type of encoding process.

The slice information data set may be transmitted in a separate network abstraction unit to the network abstraction units transmitting the encoded image. In an embodiment the slice information data set is transmitted prior to the encoded image.

In an embodiment the slice information data set further includes parameter data defining the respective coding parameters of each encoding process.

According to a fourth aspect of the invention there is provided a slice

information dataset for use with the encoding method as described hereinabove or the decoding method as described hereinabove, wherein the slice

information dataset includes information indicating the type of each different encoding process and the corresponding slice boundaries for each set of slices within the same image.

According to a fifth aspect of the invention there is provided an encoding device for encoding an image of a video sequence, the image being represented by a plurality of coding units, the encoding device being operable to apply at least two different encoding processes for encoding the image, wherein each encoding process introduces coding dependencies between encoded data associated with the respective encoding process, the encoding device comprising: slice boundary determining means for determining a set of slice boundaries for each encoding process such that the respective set of slice boundaries for each encoding process partitions the image into one or more slices for the respective encoding process, the one or more slices each including a set of coding units of the image between a pair of respective slice boundaries, wherein the coding units of one slice are encoded independently of the coding units of another of the slices of the same encoding process.

In a particular embodiment the encoding device may be operable to encode a slice information dataset for the image including slice information defining each type of encoding process and the corresponding slice boundaries within the image for each type of encoding process. In an embodiment the encoding device is operable to transmit the slice information dataset prior to the encoded image.

The encoding device may be operable to encode the slice information dataset in a separate network abstraction unit to the network abstraction units for transmitting the slices.

In an embodiment the encoding device may be operable to include the slice information dataset in a sequence parameter set of the video sequence or a picture parameter set of the image.

In an embodiment the slice boundary determining means is operable to define slices for each respective encoding process which are equal in size. In an embodiment at least one set of slice boundaries defines one or more slices equal in size to the maximum size of a transmission unit of the medium in which the video sequence is to be transmitted

In an embodiment the slice information dataset includes parameter data defining the respective coding parameters of each encoding process.

In an embodiment the encoding device is operable to apply at least one of an entropy coding and motion information prediction process. In an embodiment the slice information dataset includes information indicating if the encoding process comprises context-adaptive binary arithmetic or coding context-adaptive variable-length coding.

In an embodiment the encoding device is operable to apply intra prediction coding for pixel sample reconstruction. The slice information dataset may include information indicating the quantization parameter used to quantize the residual of a video sequence. In an embodiment the encoding device is operable to apply a loop filtering process. The slice information dataset may include information indicating the location of a loop filtering data slice where loop filtering is disabled. The slice information dataset may include adaptive loop filter data defining adaptive loop filter coefficients.

In an embodiment the slice boundary determining means is operable to define the slice boundaries according to at least one of the video content, capability of the decoder to which the video sequence is to be transmitted and characteristics of the network via which the video sequence is to be transmitted.

In an embodiment the encoding device is operable to protect the slice information data set for transmission by an error correction technique.

In an embodiment the encoding device is operable to encode one or more individual parameters subsets for each type of encoding process.

In an embodiment is operable to transmit each individual parameter subset in a separate network abstraction unit to the network abstraction units transmitting the slices.

According to a sixth aspect of the invention there is provided a decoding device for decoding a video bitstream representative of an image of a video sequence, the image being represented by a plurality of coding units and having been encoded by at least two different encoding processes, wherein each encoding process introduces coding dependencies between encoded data associated with the respective encoding process, the decoding device comprising:

decoding means for decoding the bitstream according to sets of slice

boundaries wherein each set of slice boundaries corresponds to an different decoding process and defines one or more slices encoded according to one of the different encoding processes, each slice including a set of coding units, wherein the coding units of one slice are decoded independently of the coding units of another of the slices for the same decoding process.

In a particular embodiment the decoding device comprises means for processing a received slice information dataset for each image including data identifying at least two different decoding processes for decoding the image and indicating the respective set of slice boundaries within the image for each decoding process. In an embodiment the decoding device is operable to decode two or more of the slices of a set of slices of a respective decoding process in parallel.

In an embodiment the decoding device operable to perform at least two different decoding processes in parallel.

According to a further aspect of the invention there is provided a decoding device operable to receive a slice information dataset including information indicating a plurality of different decoding processes and corresponding slice boundaries defining a set of slices within an image for each decoding process, the device being further operable to perform parallel decoding of slices according to the information in the slice information dataset.

In an embodiment the decoding device is operable to adapt the number of processing modules to be run in parallel according to the number of slices defined by the slice boundaries for each decoding process

According to a further aspect of the invention there is provided a method of decoding an image comprising receiving a slice information dataset including information indicating a plurality of different decoding processes and corresponding slice boundaries defining a set of slices within an image for each decoding process, and performing parallel decoding of slices according to the number of slices defined by the slice boundaries for each decoding process. According to a further aspect of the invention there is provided a signal carrying a slice information dataset including information indicating a type of decoding process and the corresponding slice boundaries defining a set of slices within an image for each type of decoding process.

According to another aspect of the present invention there is provided a method of encoding an image of a video sequence, the image being represented by a plurality of coding units, the encoding method comprising at least two encoding processes for encoding the image, wherein each encoding process introduces coding dependencies between encoded data associated with the encoding process concerned, the method comprising: determining for each said encoding process a set of slice boundaries which partition the image into one or more slices for the encoding process concerned, the or each said slice including a set of coding units of the image between a pair of respective slice boundaries, and encoding the coding units of slices of at least one said encoding process using the set of slice boundaries determined therefor so that the coding units of one slice of the encoding process concerned are decodable independently of the coding units of another of the slices of the encoding process concerned.

Some aspects of the invention relate to determining portions of an image that may be decoded in parallel using the same type of decoding process. Different sets of portions composing the image may be determined for different types of decoding processes, respectively. For example the image may be partitioned into two portions for one type of decoding process, and into four portions for another type of decoding process. The borders of the portions for the different decoding processes may or may not be aligned. Thus the invention further relates to partitioning a video bitstream for parallel decoding, the method comprising defining a plurality of sets of portions composing the video bitstream for a plurality of respective independent decoding processes, the bitstream being encoded such that the data of one portion of a set of portions may be decoded independently of the data of another portion of the said set of portions. Thus since the portions are coded independently they may be decoded in parallel.

Accordingly, according to a first further aspect of the invention there is provided a method of encoding an image of a video sequence, the image being represented by a plurality of coding units, the encoding method comprising at least two different encoding processes for encoding the image, wherein each encoding process introduces coding dependencies between encoded data associated with the respective encoding process, the method comprising: determining, for each encoding process, a set of image-portion boundaries wherein the respective set of image-portion boundaries partitions the image into one or more image portions for the respective encoding process, the one or more image portions each including a set of coding units of the image between a pair of respective image-portion boundaries, wherein the coding units of one image portion are encoded independently of the coding units of another of the image portions of the encoding process concerned.

According to a second further aspect of the invention there is provided a method of decoding a video bitstream representative of an image of a video sequence, the image being represented by a plurality of coding units and having been encoded by at least two different encoding processes, wherein each encoding process introduces coding dependencies between encoded data associated with the respective encoding process, the method comprising:

decoding the bitstream using different decoding processes, each corresponding to one of said encoding processes and having boundaries which partition the coding units of the image into one or more image portions for the decoding process concerned, wherein the coding units of one image portion are decoded independently of the coding units of another image portion for the decoding process concerned.

According to a third further aspect of the invention there is provided a bitstream of video data comprising: an encoded image, encoded using at least two different encoding processes wherein each encoding process introduces coding dependencies between encoded data associated with the respective encoding process and is associated with a respective set of image-portion boundaries for the respective encoding process, wherein the respective set of image-portion boundaries for each encoding process partitions the encoded image into one or more image portions, the one or more image portions each including a set of coding units of the image between a pair of respective image-portion

boundaries and wherein the coding units of one image portion are encoded independently of the coding units of another of the image portions of the same encoding process.

According to a fourth further aspect of the invention there is provided an image- portion information dataset for use with the encoding method as described hereinabove or the decoding method as described hereinabove, wherein the image-portion information dataset includes information indicating the type of each different encoding process and the corresponding image-portion boundaries for each set of slices within the same image.

Further aspects of the present invention relate to encoding and decoding devices, programs and signals corresponding to the first to fourth further aspects described above.

At least parts of the methods according to the invention may be computer implemented. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a "circuit", "module" or "system". Furthermore, the present invention may take the form of a computer program product embodied in any tangible medium of expression having computer usable program code embodied in the medium. Since the present invention can be implemented in software, the present invention can be embodied as computer readable code for provision to a programmable apparatus on any suitable carrier medium. A tangible carrier medium may comprise a storage medium such as a floppy disk, a CD-ROM, a hard disk drive, a magnetic tape device or a solid state memory device and the like. A transient carrier medium may include a signal such as an electrical signal, an electronic signal, an optical signal, an acoustic signal, a magnetic signal or an electromagnetic signal, e.g. a microwave or RF signal. Embodiments of the invention will now be described, by way of example only, and with reference to the following drawings in which:-

Figure 1 is a schematic diagram of a data communication network in which one or more embodiments of the invention may be implemented;

Figure 2 is a schematic block diagram illustrating an example of an image coding structure employed in HEVC;

Figure 3 is a schematic block diagram illustrating modules of an exemplary HEVC encoder;

Figure 4 is a schematic block diagram illustrating modules of an exemplary HEVC decoder; Figure 5 is an example of an encoded image in accordance with a first embodiment of the invention;

Figure 6 is an example of an encoded image in accordance with a second embodiment of the invention; and

Figure 7 is an example of an encoded bitstream in accordance with a further embodiment of the invention; and Figure 8 is a schematic block diagram of an encoding device or a decoding device in which embodiments of the invention may be implemented. Figure 1 illustrates a data communication system in which one or more embodiments of the invention may be implemented. The data communication system comprises a sending device, in this case a server 101 , which is operable to transmit data packets of a data stream to a receiving device, in this case a client terminal 102, via a data communication network 100. The data communication network 100 may be a Wide Area Network (WAN) or a Local Area Network (LAN). Such a network may be for example a wireless network (Wifi / 802.1 1 a or b or g), an Ethernet network, an Internet network or a mixed network composed of several different networks. In a particular embodiment of the invention the data communication system may be a digital television broadcast system in which the server 101 sends the same data content to multiple clients.

The data stream 104 provided by the server 101 may be composed of multimedia data representing video and audio data. Audio and video data streams may, in some embodiments, be captured by the server 101 using a microphone and a camera respectively. In some embodiments data streams may be stored on the server 101 or received by the server 101 from another data provider. The video and audio streams are coded by the server 101 in particular for them to be compressed for transmission.

In order to obtain a better ratio of the quality of transmitted data to quantity of transmitted data, the compression of the video data may be of motion compensation type, for example in accordance with the HEVC format or H.264/AVC format.

In order to transmit a video stream compressed using a HEVC type codec, the server 101 performs packetization which consists in embedding network abstract layer (NAL) units in network packets. A NAL unit is a data container containing a header and coded elements. For instance, one NAL unit may correspond to a slice composing a video frame. The compressed data is partitioned into packets by implementation of a packetization process and transmitted to the client 102 via the network 100 using a communication protocol, for example Real-time Transport Protocol (RTP), User Datagram Protocol (UDP) or any other type of communication protocol. Three main packetization methods may be used for each network abstract layer (NAL) unit of the bitstream. In a first case, one NAL unit size is smaller than the maximum transport unit (MTU) size corresponding to the largest packet size that can be transmitted over the network without being fragmented. In such a case, the NAL unit is embedded into a single network packet. The second case occurs when multiple entire NAL units are included in a single network packet. In a third case, one NAL unit may be too large to be transmitted in a single network packet and is thus split into several fragmented NAL units with each fragmented NAL unit being transmitted in an individual network packet. Fragmented NAL unit should be sent consecutively for decoding purposes.

The client 102 receives the packetized data and reconstitutes the NAL units from the network packet. For fragmented NAL units, the client concatenates the data from the fragmented NAL units in order to reconstruct the original NAL unit. When one of the fragmented NAL units is lost, the original NAL unit cannot be decoded and is thus discarded. The client 102 decodes the reconstructed data stream received by the network 100 and reproduces the video images on a display device and the audio data by a loud speaker. The decoding process of the client will be further described in reference to Figure 4. Figure 2 illustrates an example of an image coding structure typically employed in HEVC. According to HEVC and one of its previous predecessors, the original video sequence 201 is composed of a succession of digital images "images i". Each digital image is represented by one or more matrices, the coefficients of which represent pixels. The images 202 can be partitioned into slices 203. A slice may correspond to a part of an image or the entire image. In HEVC these slices are divided into non-overlapping Largest Coding Units (LCUs) 204 which are generally blocks of size 64 pixels x 64 pixels. Each LCU may in turn be iteratively divided into smaller variable size Coding Units (CUs) 205 using a quadtree decomposition. Coding units correspond to elementary coding elements and are constituted of two sub-units: a Prediction Unit (PU) 206 and a Transform Unit (TU) 207 which have a maximum size equal to the size of the corresponding CU. A Prediction Unit corresponds to the partition of the CU for prediction of pixels values. Each CU can be further partitioned into a maximum of 2 symmetric rectangular Partition Units 206. Transform units 207 are used to represent the elementary units that are spatially transformed by DCT (Discrete Cosine Transform). A CU can be partitioned into a TU based on a quadtree representation.

In HEVC, two kinds of slices 203 are typically used: coding slices and entropy slices. In Figure 2 frame 203 has been divided in 16 slices: CS1 , CS7 and CS13 correspond to coding slices and ES1-6, ES 8-12, and ES14-16 to entropy slices. Each coding slice starts with a slice header 208 that indicates the coding parameters of the coding slice. The slices break the coding dependencies at their boundaries which are represented in Figure 2 by dashed lines 209 and 210. Thus, the CUs of slice CS7 cannot use the CUs pixels of slice ES4 for intra prediction. An entropy slice allows syntax decoding of the bitstream independently of other entropy slices. The coding parameters of entropy slices are those of the most recent coding slice in raster scan order. To achieve such functionality, each entropy slice resets the entropy context and considers neighbourhood data outside the entropy slice as unavailable. Thus, entropy slice enables syntax information of the bitstream to be decoded in parallel. For instance, entropy slices ES2, ES3 and ES4 can be decoded in parallel. In HEVC, entropy slices and coding slices are typically represented by different NAL units. The coding parameters of the video sequence are stored in dedicated NAL units called parameter sets. In HEVC and H.264/AVC two kinds of parameter sets NAL units are employed. The first kind known as a Sequence Parameter Set (SPS) NAL unit that includes parameters that are unchanged during the entire video sequence. Typically, an SPS handles the coding profile, the size of the video frames and other parameters. The second kind known as a Picture Parameter Set (PPS) which codes the different values that may change from one image to another.

Figure 3 schematically illustrates an exemplary HEVC video encoder 30. Each frame of the original video sequence 201 is first divided into a grid of coding units (CU) during stage 301 and coding slices and entropy slices are defined. In general, two methods define slice boundaries by either defining a given number of CUs per slices (entropy or coding slices) or a given number of bytes per slice. The subdivision of an LCU into CUs and the partitioning of a CU into TUs and PUs are determined based on a rate distortion criterion. Each PU of the CU being processed is predicted spatially by an "Intra" predictor 317, or temporally by an "Inter" predictor 318. Each predictor is a block of pixels issued from the same image or another image, from which a difference block (or "residual") is derived. By identifying a predictor block and coding the residual, it is possible to reduce the actual quantity of information to be encoded.

The encoded images can be of two types: temporal predicted images which can be either predicted from one or more reference images in one direction are called P-frames or predicted from at least two reference frames in two directions (forward and backward) are called B-frames; and non-temporal predicted frames called Intra frames or l-frames. In l-frames, only Intra prediction is considered for coding CUs/PUs. In P-frames and B-frames, Intra and Inter prediction are considered for coding CUs/PUs. In the "Intra" prediction processing module 317, the current block is predicted by means of an "Intra" predictor which corresponds to a block of pixels constructed from the information of the current image already encoded. The module 302 determines an angular prediction direction that is used to predict pixels of a current PU to encode from neighbouring PUs pixels. In HEVC, up to 34 directions can be considered. A residual PU is obtained by computing the difference between the intra predicted PU and current PU of pixels. An intra- predicted PU therefore comprises a prediction direction with a residual. The coding of the intra prediction direction is partly inferred from the prediction direction of neighbouring prediction units. This inferring process 303 of prediction direction enables the coding rate of the intra prediction direction mode to be reduced. The Intra prediction processing module 317 also uses the spatial dependencies of the frame for predicting the pixels and for inferring the intra prediction direction of the prediction unit.

With regard to the second processing module 318 that is "Inter" coding, two prediction types are possible. Mono-prediction (P-type) entails predicting the PU by referring to one reference area from one reference picture. Bi-prediction (B- type) entails predicting the PU by referring to two referenceareas from one or two reference pictures. In HEVC, B-type frames have been generalized and replace P-type frames which now predict the PU by referring to two reference areas in one reference picture. An estimation of motion 304 between the current PU and reference images 315 is made in order to identify, in one or several of these reference images, one (for P-type) or several (for B-type) areas of pixels to use them as predictors of this current PU. In the case where several areas predictors are used (B-type), they are merged to generate one single prediction . The reference images are images in the video sequence that have already been coded and then reconstructed (by decoding). The reference area is identified in the reference frame by a motion vector that is equal to the displacement between the PU in current frame and the reference area. The next stage 305 of the inter prediction process involves computing the difference between the prediction area and the current PU. This difference is the residual of the inter predicted PU. At the end of the inter prediction process the current PU is composed of one motion vector and a residual. By virtue of spatial dependencies of movement between neighbouring PUs, HEVC provides a method to predict the motion vectors of each PU. Several motion vector predictors are employed: typically, the motion vector of the PU localized on the top of, the left of or the top left corner of the current PU are a first set of spatial predictors. One temporal motion vector candidate is also used that is the one of the collocated PU (i.e. the PU at the same coordinate) in a reference frame. The coder then removes predictors that are equal within the set of candidates. It selects one of the predictors based on a criterion that minimizes the difference between the MV predictor and that of the current PU. In HEVC, this process is referred to as Advanced Motion Vector Prediction (AMVP). Finally, the current PU's motion vector is coded 306 with an index that identifies the predictor within the set of candidates and a MV difference MVD of PU's MV with the selected MV candidate. The inter prediction processing module also relies on spatial dependencies between motion information of prediction units to increase the compression ratio of inter predicted coding units. For entropy slices, an AMVP process neighbouring PUs are considered as unavailable if they do not belong to the same slice.

These two types of codings thus supply several texture residuals (the difference between the current PU and the predictor), which are compared in a module 316 for selecting the best coding mode.

The residual obtained at the end of an inter or intra prediction process is then transformed in module 307. The transform applies to a Transform Unit TU that is included in a CU. A TU can be further split into smaller TUs using a so-called Residual QuadTree (RQT) decomposition 207. In HEVC, generally 2 or 3 levels of decompositions are used and authorized transform sizes are from 32x32, 16x16, 8x8 and 4x4. The transform basis is derived from a discrete cosine transform DCT.

The residual transformed coefficients are then quantized 308. The coefficients of the quantized transformed residual are then coded by means of an entropy coding process 309 and then inserted into the compressed bit stream 310. Coding syntax elements are also coded by entropy encoding 309. This processing module uses spatial dependencies between syntax elements to increase the coding efficiency.

In order to calculate the "Intra" predictors or to make an estimation of the motion for the "Inter" predictors, the encoder performs a decoding of the PUs already encoded by means of a so-called "decoding" loop 31 1 , 312, 313, 314, 315. This decoding loop makes it possible to reconstruct the PUs and images from the quantized transformed residuals.

Thus the quantized transformed residual is dequantized 31 1 by applying the inverse quantization to that provided at quantization step 308 and reconstructed 312 by applying the inverse transform to that of the step 307.

If the residual comes from an "Intra" coding process 317, the used "Intra" predictor is added to this residual in order to recover a reconstructed PU corresponding to the original PU modified by the losses resulting from a transformation with loss, for example in this case the quantization operations.

If on the other hand the residual comes from an "Inter" coding 318, the areas pointed to by the current motion vectors (these areas belong to the reference images 315 referred by the current image indices) are merged then added to this decoded residual. In this way the original PU is modified by the losses resulting from the quantization operations. A final loop filter processing module 319 is applied to the reconstructed signal in order to reduce the effects created by heavy quantization of the residuals obtained and to improve the signal quality. The loop filter processing module comprises two steps, a "deblocking" filter and a linear filtering. The deblocking filter 313 smoothes the borders between the PUs in order to visually attenuate the high frequencies created by the coding. Such a filter being known to the skilled person, it will not be described in any further detail here. The linear filtering 314 Adaptive Loop Filter (ALF) further improves the signal using filter coefficients adaptively determined. The coefficients of the filter are computed in order to minimize the sum of square difference (SSD) with the original image. The coefficients of the filters are coded and transmitted in one header of the bitstream typically a picture or slice header. The filter 319 is thus applied to an image when all the PUs of pixels of the image have been decoded. The filtering process is performed on a frame by frame basis and uses several pixels around the pixel to be filtered. This processing module 319 also uses spatial dependencies between pixels of the frame.

The filtered images, also known as reconstructed images, are then stored as reference images 315 in order to allow the subsequent "Inter" predictions taking place during the compression of the subsequent images of the current video sequence.

In the context of HEVC, it is possible to use several reference images 315 for the estimation and motion compensation of the current image. In other words, the motion estimation is carried out on N images. Thus the best "Inter" predictors of the current PU, for the motion compensation, are selected in some of the multiple reference images. Consequently two adjoining PUs may have two predictor PUs that come from two distinct reference images. This is particularly the reason why, in the compressed bit stream, the index of the reference image (in addition to the motion vector) used for the predictor area is indicated. The use of multiple reference images, in particular where the number of reference images is limited to 4, is both a tool for resisting errors and a tool for improving the compression efficacy. The encoder is embedded in a server such as the server 101 of Figure 1 for streaming purposes and therefore slices may be employed for error resilience. For instance, the number of coding slices is set to a predetermined number (e.g. 8 slices per frame for 1920x1080 resolution) in order to limit the error propagation within a frame. Entropy slices are used to enable parallel processing on the decoder side. The coding process presented previously is also adapted to respect coding and entropy slices characteristics as described with reference to Figure 2. The resulting bitstream 310 is also composed of a set of NAL units corresponding to parameter sets, coding slices and entropy slices.

Figure 4 is a block diagram of an exemplary HEVC type video decoder 40. The decoder 40 receives as an input a bit stream 310 corresponding to a video sequence 201 compressed by an encoder of the HEVC type, such as the one in Figure 3. During the decoding process, the bit stream 310 is first of all parsed with help of the entropy decoding module 401. This processing module uses the previously entropy decoded elements to decode the encoded data. It decodes in particular the parameter sets of the video sequence to initialize the decoder and also decodes LCUs of each video frame. Each NAL unit that corresponds to coding slices or entropy slices is then decoded. The parsing process that comprises entropy decoding 401 , decode prediction direction 402 and decode motion information 404 stages can be done in parallel for each slice but PL ⁾ prediction processes module 405 and 403 and loop filter module are preferably sequential to avoid issues of neighbouring data availability. The partition of the LCU is parsed and CU , PU and TU subdivision are identified. The decoder successively processes each CU by intra (407) or inter (406) processing modules, inverse quantization and inverse transform modules and finally loop filter processing module (319).

The "Inter" or "Intra" coding mode for the current block is parsed from the bit stream 310 with help of the parsing process module 401. Depending on the coding mode, either intra prediction processing module 407 or inter prediction processing module 406 is employed. If the coding mode of the current block is "Intra" type, the prediction direction is extracted from the bit stream and decoded with help of neighbours' prediction direction during stage 404 of intra prediction processing module 407. The intra predicted block is then computed (403) with the decoded prediction direction and the already decoded pixels at the boundaries of current PU. The residual associated with the current block is recovered from the bit stream 401 and then entropy decoded. If the coding mode of the current PU indicates that this PU is of "Inter" type, the motion information is extracted from the bit stream 401 and decoded (404). AMVP process is performed during step 404. Motion information of neighbours PU already decoded are also used to compute the motion vector of current PU. This motion vector is used in the reverse motion compensation module 405 in order to determine the "Inter" predictor PU contained in the reference images 315 of the decoder 40. In a similar manner to the encoder, these reference images 315 are composed of images that precede in decoding order the image currently being decoded and that are reconstructed from the bit stream (and therefore decoded previously).

The next decoding step consists in decoding the residual block that has been transmitted in the bitstream. The parsing module 401 extracts the residual coefficients from the bitstream and performs successively the inverse quantization (311) and inverse transform (312) to obtain the residual PU. This residual PU is added to the predicted PU obtained at output of intra or inter processing module. At the end of the decoding of all the PUs of the current image, the loop filter processing module 319 is used to eliminate the block effects and improve the signal quality in order to obtain the reference images 315. As done at the encoder, this processing module employs the deblocking filter 313 and then the ALF 314.

The images thus decoded constitute the output video signal 408 of the decoder, which can then be displayed and used. The decoder in this example comprises three main processing modules:

1. Syntax decoding module (401 ; 402; 404)

2. Pixel reconstruction module (405; 403; 31 1 ; 312)

3. Loop filter processing module (319) Figure 5 represents an example of an image encoded in accordance with a first embodiment of the invention. In the first embodiment of the invention the partitioning module 301 of the encoder 30 of Figure 3 is modified as follows: the encoder uses a single NAL unit containing some frame encoded data per frame 202 and transmits a single header 500 at the beginning of the frame. Header 500 includes coding parameters, typically slice type, inter or intra prediction parameters and loop filter parameters. A new NAL unit 504 containing a slice information data set is created by the encoder and may be referred to as a slice parameter set NAL unit. The slice information data set includes data representative of a set of parameters indicating the different coding dependencies boundaries (501 , 502 and 503) of the bitstream for each of the processing modules of the encoder, and identifying the processing type of each processing module. The encoding processes of the processing modules may include, for example, syntax encoding, pixel sample reconstruction and loop filtering. The set of coding units (CUs) between two coding boundaries is generally referred to as a "processing slice". The pseudo-syntax proposed below in Table 1 represents an example of the syntax of a new NAL unit 504 referred to as a Slice Parameter Set NAL unit.

In some embodiments of the invention, the slice parameter set may include data defining the coding parameters, in which case the image may be transmitted without slice header 500.

In embodiments of the invention the slice information data set is encoded by the encoder. For example, each syntax element presented in bold in Table 1 can be binary coded preferably by a variable length coding method . It will be appreciated however that the invention is not limited to this particular binary coding method. The semantics of each coded item of the example of Table 1 are presented as follows:

• num_processing_slice_type: An integer value greater than 0 which is equal to the number of different types of processing slice in the image

• processing_slice_type: An integer value that represents the type of the processing slice. The list of processing slice types is predetermined. The number of processing_slice_type syntax values coded in the bistream is equal to num_processing_slices_types

· num_slice: For each processing slice type, the number of slices in an image is indicated by num_slice syntax element. The number of num_slice syntax values coded in the bistream is equal to num_processing_slices_types

• slicejndex: For each processing slice type, it is the index of the first CU of each processing slice of a specific type. The index value could be a byte address or a CU address. In the preferred embodiment, processing slices are defined at LCU level and thus slicejndex is indicating one largest coding unit index.

• rbsp_trailing_bits() is used to align the end of the NAL unit on byte.

In another embodiment, the slice information data set may be sent in an optional Supplementary Enhancement Information (SEI) NAL unit. SEI NAL units embeds proprietary information that are not normalized. A decoder that does not support SEI NAL unit can then discard them. The advantage of such a method is that the bitstream may be left unchanged and remain compatible with prior standards (e.g. H.264/AVC)

In the first embodiment, three types of coding dependency boundaries corresponding to three different types of encoding processes are considered:

1. The syntax coding boundaries (503) that enable the coding dependencies for entropy coding (309) and prediction mode coding (303 and 306) sub processes to be limited. CUs that are separated with such a boundary cannot use their neighbor's CU information to code the syntax values of current CU. The set of CUs that are bounded by two syntax coding boundaries are now referred as syntax slice.

2. Pixel samples reconstruction boundaries (502) that delimit spatially the de pe nd en cies with nei g h bor C U s for cod i n g of pixels d ata

(302;304;305;307;308). The encoding process of stage 302, 304 and 305 cannot use pixel data of a neighbor CU when they are separated by these boundaries. The set of CUs that are bounded by two pixel sample boundaries are now referred as reconstruction slice.

3. Loop filtering boundaries (501) that limit the loop filter process (319) that is composed of the deblocking filter (314) and ALF process (313). These two filters cannot use pixels across a boundary when filtering one pixel localized at the vicinity of such boundaries. The set of CUs that are bounded by loop filtering boundaries may be referred to hereafter as loop filter slice. Syntax slices, reconstruction slices and loop filter slices may be referred to in following section by the generic term "processing slices".

The encoder encodes the locations of the processing slice boundaries in the NAL unit 504. The pseudo-syntax proposed below in Table 2 represents an example of the syntax of the new NAL unit 504 containing the slice information data set referred to as Slice Parameter Set NAL unit (SIPS NALU).

slice_parameter_set_nal_rbsp ( ) {

num_syntax_slices_minus1

for ( i = 0; i < num_syntax_slices_minusl ; i++)

syntax_s1ice_del a_byte_index [i ]

num_recons_s1ice_minus1

for ( i = 0; i < num_recons_slices_minusl ; i++)

recons_slice_delta_index_in_CU [i]

num_loop_fi1ter_s1ices_minus1

for ( i = 0; i < num_loop_filter_slices_minusl ; i++) loop_filter_slice_delta_index_in_CU [ i ]

rbsp_trailing_bits ( )

Table 2

Table 2 may be included in addition to Table 1 in the Slice Parameter Set NAL unit.

The semantics of each coded item of the example is presented as follows:

• num_syntax_slices_minus1 is a positive integer value that corresponds to the number of syntax slices minus 1 used in the current frame. A value of 0 for this item means that no additional syntax coding boundaries are defined in the frame than the one at the beginning of the picture. For any value greater or equal to 1 , first CUs of the syntax slices are defined by a set of syntax_slice_delta_byte_index values

• syntax_slice_delta_byte_index is a byte offset in the frame that indicates a CU that is the first one in raster scan order of new set of coding units for which syntax is independently coded. There are as many

syntax_slice_delta_byte_index values as indicated by

num_syntax_slices_minus1 syntax element. num_recons_slice_minus1 is a positive integer value that corresponds to the number of reconstruction slices minus 1 used in the current frame. A value of 0 for this item means that no additional sample reconstruction boundaries are defined in the frame than the one at the beginning of the picture. For any value greater or equal to 1 , the pixel coding boundaries are defined by a set of recons_slice_delta_index_in_CU values.

recons_slice_delta_index_in_CU is a CU address offset in the frame that indicates one CU that is the first one in raster scan order of

reconstruction slices. There are as many

recons_slice_delta_index_in_CU values as indicated by

num_recons_slice_minus1 syntax element.

num_loop_filter_slices_minus1 is a positive integer value that

corresponds to the number of loop filter slices minus 1 used in the current frame. A value of 0 for this item means that no additional loop filtering boundaries are defined in the frame than the one at the beginning of the picture. For any value greater or equal to 1 , the syntax coding boundaries are defined by a set of

loop_filter_slice_delta_index_in_CU values

loop_filter_slice_delta_index_in_CU is a CU address offset in the frame that indicates one CU that is the first one in raster scan order loop filter slice. There are as many loop_filter_slice_delta_index_in_CU values as indicated by num loop filter slices minusl syntax element.

The first CU address of a syntax slice is computed relative to the first CU of previous syntax slices in raster scan order. To decode the byte address of the first CU of the I ^th syntax slice, the decoder has to sum the first i syntax_slice_delta_byte_index values. Same principle applies for recons_slice_delta_index_in_CU and loop_filter_slice_delta_index_in_CU syntax elements. This enables the amount of bits needed to code the processing slices to be reduced. The encoder creates syntax decoding boundaries with a fixed number of bytes between two boundaries, i.e. a fixed number of bytes can be used for syntax slices. In a particular embodiment of the invention for example, the pre- determined n umber of bytes may be eq ual to th e maxi m u m size of a transmission unit (MTU) i.e. a network packet. The syntax boundaries may also be defined in terms of a fixed number of CUs between two boundaries.

Sample reconstruction boundaries may be defined such that a predetermined number of CUs is maintained between two boundaries. Each reconstruction slice may in some embodiments be defined to enclose the same number of CUs or in other embodiments to have different number of CUs. For loop filter boundaries, the frame may be divided in a predetermined number of areas that have the same surface in CU. Loop filter slices may also be defined by a fixed number of CUs.

It will be appreciated that slices corresponding to the same processing type may be defined by the corresponding slice boundaries to each be of the same size or to be of different sizes.

As a main difference with the prior art, this process enables coding boundaries to be defined advantageously for several processing modules (also called encoding processes) independently. The boundaries can be adapted to the application or the characteristics of the corresponding processing modules. For instance, the syntax decoding boundaries may be defined to achieve a high level of error resilience in case of packet loss during network transmission: the syntax coding boundaries matches the requirement of the network characteristics. Sample reconstruction boundaries are defined independently of syntax coding boundaries which permit to save bitrate. Indeed sample reconstruction boundaries are generally drastically increasing the bitrate of a video sequence. Avoiding the use of synchronised syntax coding and sample reconstruction boundaries enables the bitrate to be saved while maintaining the interest of creating coding boundaries for limiting error propagation. Defining independent boundaries for loop filter process permits to adapt the filtering to the content of the video sequence.

In one embodiment, other syntax elements are added to the slice parameter set in addition to the processing slices definition. These elements may define coding parameters that applies only for one type of processing slice. These parameters must be used to refine the coding parameter of one type of processing slice. For instance, for syntax coding, the entropy coding method Context-adaptive binary arithmetic coding (CABAC) or Context-adaptive variable-length coding (CAVLC) can be defined. This enables adaptive selection of the optimal entropy coding method for specific areas of the bitstream. Another usage is to adaptively disable loop filters for specific areas between two loop filter coding boundaries. The encoder could deactivate loop filters for areas where loop filters are useless or where no visual improvement is provided. Different ALF coefficients can also be defined for each of the loop filter slices. For sample reconstruction process, specific coding parameters such as the weighted prediction values can be used.

The slice parameter sets NAL (SIPS) unit including the slice information dataset is sent prior the slice NAL units. One coding dependencies NAL unit is sent for each access unit to define the coding boundaries for each frame. In one embodiment, if the coding dependencies boundaries are pre-defined for the whole video sequence the coding boundaries are defined in Sequence Parameter Sets. If the coding boundaries change for each frame, the SIPS NAL unit can be merged with Picture Parameter Set.

The decoder 40 could advantageously use this SIPS NAL unit to improve its decoding process. Indeed, for each access unit (i.e. coded data corresponding to one picture), the decoder starts the decoding process of SIPS NALU. From the extracted parameters, the decoder is able to run in parallel syntax decoding processes. Each syntax decoding process starts at the byte offset indicated with help of syntax_slice_delta_byte_index values of SIPS NAL unit. Similarly, pixel reconstruction and loop filtering processes are also parallelized for each corresponding processing slice.

Another advantage of the embodiment of the invention resides in the fact that the decoder could allocate its decoding resources in function of the SIPS NAL unit parameters. Indeed, the number of coding boundaries for each processing module is defined in this NAL unit which is sent prior to the coded data corresponding to the frame. Thus, the decoder can adapt the number of processing modules that could be run in parallel. For instance, considering that the SIPS NAL unit indicates 4 independent syntax coding areas (i.e. 4 syntax slices), 2 independent pixel reconstruction areas (i.e. 2 reconstruction slices) and 6 loop filtering areas (i.e. 6 loop filter slices) and the decoder is able to run 6 threads in parallel. As syntax decoding is often the most CPU consuming process, the decoder allocates 4 threads to decode the syntax slices and uses 2 threads for reconstruction slices and loop filter slices.

Figure 6 schematically illustrates an image coded in accordance with a second embodiment of the invention. In the same manner as the first embodiment, the coding boundaries are determined by block partitioning module 301. In this embodiment the syntax slices 602 and 603are represented in the bitstream by NAL units. NAL units convey the coded data and replace usual slices (as defined in AVC for instance) and entropy slices. A single type of slice, such as a syntax slice is thus transmitted as an NAL unit. The advantage of this method is that the packetization process of the server can be simplified since syntax slices are represented in independent NAL units. Only syntax elements that are related to syntax coding process parameters are presented in syntax slices headers. Moreover, syntax elements that are related to syntax coding process parameters and that change from one slice to another and which are stored in another NAL unit are represented in the syntax slice header. Typically, the entropy_code_mode_flag defined in PPS for H.264/AVC is moved to the syntax slice header. In addition, Picture Order Count (POC) syntax elements, quantizer parameters and a deblocking filter flag which are classically defined at slice level for H.264/AVC and HEVC are added to the slice parameter information SIPS NAL unit.

Coding syntax of syntax slices that may be used for this embodiment is as follows: syntax_slice_layer_rbsp ( ) {

entropy_coding_mode_flag

firs _lctb_in_slice

slice_data ( )

rbsp trailing bits()

Table 3

Entropy_coding_mode_flag is a flag that states if Context-adaptive binary arithmetic coding (CABAC) or Context-adaptive variable-length coding (CAVLC) coding should be employed for decoding current slice

First_lctb_in_slice indicates the address of the first CU of the slice in the image

Slice_data() represent the coding data of the slice (similar to HEVC or

AVC)

In one embodiment, an entropy_coding_mode_flag is fixed for the entire video sequence - in such a case it can be defined in SPS or PPS NAL unit instead of or in addition to in the slice NAL unit.

Other coding boundaries (501 ; 502) are still coded in SIPS NAL unit. Coding parameters dedicated to one type of processing (either pixel reconstruction or deblocking filter processes) are stored in addition to the definition of processing slice boundaries. These parameters can be defined in common for all processing slices or independently for each one. For instance, the pseudo code below in Table 4 represents the syntax of one slice parameter set (604) of an embodiment of the present invention.

frame_type refers to the type of slices. The slices may be either Inter or Inter. In most situations the prediction type of entropy slices are defined for each frame. Every processing slice of the image has the same type. For this reason slice_type is defined for one picture or image.

frame_num is a value that counts the number of frames that have been used as a reference from the beginning of the sequence. It allows the detection of reference frame loss. This value does not change from one processing slice to another; therefore it is defined for all processing slices of the frame.

· pic_order_cnt is a value that represents the index of the frame in decoding order. In HEVC, this value is defined for each HEVC slice. In this embodiment of the invention, this value is defined for all processing slices of the frame since the value remains unchanged between two coding areas.

recons_slice_qp_delta is the quantization parameter value that is used to quantize the residual of a sequence. The value of this parameter could be adapted as a function of the high frequencies of each reconstruction slice. One syntax element is thus coded for each reconstruction slice.

Pred_weight_table() refers to the coding syntax of weighted prediction coefficients for inter prediction process. This process which is clear to the skilled person in the art is not detailed further in this document. Coding syntax similar to AVC or HEVC may be used.

ref_picJist_reordering() refers to the coding syntax of commands to reorder the reference pictures in the reference picture buffer 315. This syntax is not detailed further in this document and syntax similar to AVC or HEVC is used.

disable_deblocking_filter_idc is a flag that disables the deblocking filter. The value of this parameter could be adapted as a function of each loop filter slice. One syntax element is thus coded for each loop filter slice.

alf_param() refers to the coding syntax of adaptive loop filter coefficients. This syntax is not detailed further in this document and syntax similar to HEVC is used.

The advantage of including coding parameters of processing slices in a slice parameter set NALU is that this information is not duplicated in slices. A number of bits are thus saved with this method and the bitrate of the resulting bitstream is lower in proposed embodiments of the invention. Moreover, the decoder is able to initiate its coding process modules before decoding the slice data. The CPU resource shared between the processing units could be adapted based on these coding parameters.

In a streaming scenario, network packet transmission may protected by an error correction technique, for example by forward error correction codes (FEC). The principle of FEC is to generate new FEC packets from a predetermined number of network packets. A FEC packet is capable of correcting for loss of one packet if all the other packets among the predetermined packets are correctly received. FEC packets increase the bitrate of the sent data and may be used to protect the most important data that may typically include the parameters of AVC's coding slice header. Another advantage of including coding parameters of processing slices in a slice parameter set NALU is that the size of the parameters is negligible in comparison to AVC coding slice data. The FEC data bitrate required to protect these parameters is thus reduced in comparison of AVC. Indeed, the FEC packets are not required to also protect the slice data since the slice parameter data and the slice data are embedded in separate NAL units.

The previous embodiments could be improved by creating parameter set NAL unit types for each type of processing slice. Each processing slice may then use an identifier value to retrieve the appropriate processing slice parameter set NAL unit. The new processing slice parameter sets NALU enables the parameters of each processing slice of the same type to be gathered in a single NAL unit. Table 5 presents an example of the coding syntax for the reconstruction slice parameter set NAL unit:

Recons_s l ice_paramete r_set_rbsp

Recons_slps_id

recons_s1ice_qp_de11a

pred_we ight_table ( )

re f_pic_list_reorde ring ( )

rbsp_trail ing_bits ( ) Table 5 recons_slps_id is an integer value indentifying a Reconstruction Slice Parameter Set (RSIPS). This identifier can be unique for each RSIPS of one access unit of the video sequence.

recons_slice_qp_delta is a value representing the quantization parameter.

the remaining syntax elements have already been described in previous embodiments The following lines in Table 6 present an example of the coding syntax of the loop filter slice parameter sets.

Loop filter slice parameter set rbsp ( ) {

If spls id

disable deblocking filter_idc

alf param list()

rbsp trailing bits ( )

}

Table 6

lf_slps_id is an integer value that identifies a Loop filter Slice Parameter Set (LSIPS). This identifier can be unique for each ISIPS NAL unit of one access unit of the video sequence.

· disable_deblocking_filter_idc is a flag that permits to disable the deblocking filter for one loop filter slice.

the remaining syntax elements have already been described in previous embodiments

The syntax of the SIPS NAL unit determines the different processing slices of one frame and for each processing slice, one processing slice' parameters set identifier is coded. For example, in coding syntax below, recons_slps_id (respectively lf_slps_id) is an identifier value that refers to one RSIPS (reconstruction SLPS) NAL unit (respectively LSIPS (loop filter SLPS) NAL unit). One identifier is specified for each processing slice.

slice_parameter_set_nal_rbsp ( ) {

num_recons_slice_minusl

for ( i = 0; i < num_recons_slices_minusl ; i++)

{

recons_s1ice_delta_index_in_CU [i]

recons_slps_id [i ]

}

num_loop_fi1 er_slices_minus1

for ( i = 0; i < num loop filter slices minusl; i++) {

1oop_fi1ter_s1ice_delta_index_in_CU [i ]

lf_slps_id [i]

}

rbsp_trailing_bits ( )

}

Table 7

This embodiment enables several processing parameters to be obtained for each processing slice unit that can be adapted during the coding of the sequence. Figure 7 schematically represents NAL units that constitute the encoded bitstream in accordance with the third embodiment. The encoded bitstream begins with Sequence Parameter Set (702) and Picture Parameter Set (703) NAL units. Then, the encoded frame 700 starts with a slice parameter information dataset SIPS NALU 704. This NAL unit is sent prior to any NAL slice units of the frame. Then a processing slice parameter set NALU dedicated to processing parameters for reconstruction slices (705) and a processing slice parameter set NALU dedicated to processing parameters for Loop filter slices 706 are transmitted. The order of these two NALUs is not imposed and the NALU for the Loop filter slice parameter set 706 may be transmitted prior to the reconstruction slice parameter set NALU 705. For image 700, slice processing parameters are the same for the entire frame. Thus no other processing slice parameter sets NAL units are transmitted.

For new frame 701 , the encoder defines new SIPS 708 and slice processing parameter sets NALU 709 and 710 at the beginning of the new frame. The identifier of NALU 709 and 710 is reset to 0 since a new frame has been started. Then the encoder encodes coding units CUs of frame 701. After having encoded a first set of CUs, the encoder handles a new reconstruction slice which has different processing parameters. Since, the first CU of this new reconstruction slice is coded in the second slice, the new parameter sets of the reconstruction slice are transmitted prior to slice 713 in NALU 712. This example illustrates one property of Slice Processing Parameter Sets: the availability of a Slice Processing Parameter Set should be prior to the coding of the first CU that refers to this Parameter Sets. For instance, the frame 701 is divided into three loop filter slices that have different processing parameters. The CUs of first loop filter slice are embedded in slices 711 and 713. The CUs of second and third loop filter slices are contained in the last slice 716. Thus, Loop Filter Slice Parameter Sets of the two last loop filter slices should be available prior to the slice 716. In the example of Figure 7, the encoder chooses to send Loop Filter Slice Parameter Sets 714 and 715 just before the slice 716 but alternatively they could have been sent just after LSIPS 710.

In the preceding embodiments three types of encoding processes have been considered: a syntax coding process, a pixel sample reconstruction coding process and a loop filter process. HEVC syntax coding processes are typically composed of entropy coding and motion vector prediction processes. This is due to the fact that the entropy process uses decoded motion vectors values of neighbouring CUs to code one codeword. This limitation requires that the two processes are considered as one process. In another embodiment it may be considered that the entropy coding processes and motion vector coding are not dependent. For example, an entropy coder may use simple coded motion information such as reference picture index to compute codewords. In such a case, it will be appreciated that the syntax process can be split into two independent processes and new processing slices can be defined accordingly.

Figure 8 illustrates a block diagram of a receiving device such as client terminal 102 of Figure 1 or transmitting device, such as server 101 of Figure 1 which may be adapted to implement embodiments of the invention. The device 800 comprises a central processing unit (CPU) 801 capable of executing instructions from program ROM 803 on powering up of the device 800, and instructions relating to a software application from a main memory 802 after powering up of the device 800. The main memory 802 may be for example of a Random Access Memory (RAM) type which functions as a working area of the CPU 801 , and the memory capacity thereof can be expanded by an optional RAM connected to an expansion port (not illustrated). Instructions relating to the software application may be loaded into the main memory 802 from a hard-disc (HD) 806 or the program ROM 803 for example. Such a software application, when executed by the CPU 801 , causes the steps of the method of embodiments of the invention to be performed on the device. The device 800 further includes a network interface 804 enabling connection of the device 800 to the communication network. The software application when executed by the CPU is adapted to receive data streams through the network interface 804 from other devices connected to the communication network in the case of a receiving device and to transmit data streams through the network interface 804 in the case of a transmitting device.

The device 800 further comprises a user interface 805 to display information to, and/or receive inputs from, a user. Although the present invention has been described hereinabove with reference to specific embodiments, the present invention is not limited to the specific embodiments, and modifications will be apparent to a skilled person in the art which lie within the scope of the present invention.

For example, while the foregoing embodiments have been described with respect to three types of encoding/decoding processes it will be appreciated that the invention may be adapted any number of types of encoding/decoding processes, and to encoding processes of different types to those described above. Many further modifications and variations will suggest themselves to those versed in the art upon making reference to the foregoing illustrative embodiments, which are given by way of example only and which are not intended to limit the scope of the invention, that being determined solely by the appended claims. In particular the different features from different embodiments may be interchanged, where appropriate.

In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. The mere fact that different features are recited in mutually different dependent claims does not indicate that a combination of these features cannot be advantageously used.