IDENTIFYING AND SELECTING DATA IMAGES IN A VIDEO STREAM

[0001] The disclosed subject matter relates to systems, methods, and media for identifying and selecting data images in a video stream.

[0002] In various domains, data is shared between remote participants using existing video channels. To achieve this, data is captured, and then encoded using a standard video encoder as if it was natural video. The remote parry receives the video and the data (e.g., as data images making up a data video) using a video decoder. However, in these scenarios, because the data is a real-time video stream, a user cannot easily browse the data (for instance, slides in a presentation), review data that was previously presented (for instance, when arriving late for a video conference), distribute data presented (for instance, during a video conference) after it was shared, etc.

Summary

[0003] Systems, methods, and media for identifying and selecting data images in a video stream are provided. In accordance with some embodiments, systems for identifying and selecting data images in a video stream are provided, the systems comprising: at least one processor programmed to: receive a video bit stream representing a plurality of images; identify sets of data images in the plurality of images, wherein each of the data images provides an image representation of data and each of the data images in any of the sets of data images corresponds to a single data image; and select at least one best data image from the sets of data images.

[0004] In accordance with some embodiments, methods for identifying and selecting data images in a video stream are provided, the methods comprising: receiving a video bit stream representing a plurality of images; identifying, using a processor programmed to do so, sets of data images in the plurality of images, wherein each of the data images provides an image representation of data and each of the data images in any of the sets of data images corresponds to a single data image; and selecting at least one best data image from the sets of data images using a processor programmed to do so.

[0005] In accordance with some embodiments, computer-readable medium containing computer-executable instructions that, when executed by a processor, cause the processor to perform a method for identifying and selecting data images in a video stream are provided, the method comprising: receiving a video bit stream representing a plurality of images; identifying sets of data images in the plurality of images, wherein each of the data images provides an image representation of data and each of the data images in any of the sets of data images corresponds to a single data image; and selecting at least one best data image from the sets of data images.

Brief Description of the Drawings

[0006] FIG. 1 is a diagram of hardware that can be used in some embodiments.

10007] FIG. 2 is a diagram of a process for processing a video bit stream in accordance with some embodiments.

[0008] FIG. 3 is a diagram of a process for identifying sets of data images in accordance with some embodiments.

[0009] FIG. 4 is a diagram of a table showing filter passes and false positive rates, false negative rates, and complexities associated with those filter passes in accordance with some embodiments.

[0010] FIG. 5 is a diagram of a process for selecting best data images in sets of data images in accordance with some embodiments.

Detailed Description

[001 1] In accordance with various embodiments, mechanisms for identifying and selecting data images in a video stream are provided. These mechanisms can be used in a variety of applications such as to allow for online browsing of data images, to allow for late arrivals in a video conference to receive data that was previously shown, to allow for later distribution of data images provided in a video conference, and to allow browsing through a recorded video conference call using data images as an index.

[0012] Turning to FIG. 1, an example of hardware 100 that can be used i some embodiments is illustrated. As shown, data 102 can be provided to a video encoder 106 using a camera 104 or data 102 can be provided to a video encoder 108 directly. The data can be any suitable data and may be in the form of a visual presentation (e.g., such as a POWERPOINT presentation). Camera 104 can be any suitable camera, such as a video camera. Video encoders 106 and 108 can be any suitable hardware and/or software video encoders and may utilize any suitable video encoding standard(s) and/or coder(s)/decoder(s) (codec(s)) for creating a video signal and/or compressing the signal. In some embodiments, camera 104 and/or video encoders 106 or 108 can be part of a video conferencing endpoint

[0013] After the data is encoded, a bit stream may be provided by the video encoder. This bit stream may be transmitted via a network 1 10. Any suitable network and/or combination of networks may be used as network 1 10 in some embodiments. For example, network 1 10 may include the Internet, a wired network, a wireless network, a local area network, a wide area network, a telephone network, a cable network, a satellite network, a fiber optic network, etc. In some embodiments, network 110 can include any suitable equipment such as routers, firewalls, servers, proxy servers, multipoint conferencing units, gateways, etc.

[0014] The bit stream may be provided to a network appliance 1 12 and/or an endpoint 122 in some embodiments (e.g., via network 110, directly from encoder 106 or 108, etc.).

Network appliance 1 12 may be any suitable device or part of any suitable device. For example, appliance 1 12 may be a server. Appliance 112 may include a video decoder 114 and an image identifier and selector 116. Endpoint 122 may be any suitable device or part of any suitable device. For example, endpoint 122 may be a computer that is configured as a video conferencing endpoint. Endpoint 122 may include a video decoder 124 and an image identifier and selector 126. Endpoint 122 may also include endpoint components 128 and a data image repository 130.

[0015] Video decoders 114 and 124 may be any suitable hardware and/or software video decoder for decoding the video bit stream into a series of video images. Video decoders 114 and 124 may utilize any suitable video encoding standard(s) and/or coder s) decoder(s) (codec(s)). For example, the video decoders may include a decoder that matches the encoder in video encoder 106 or 108.

[0016] Image identifier and selector 11 and 126 may be any suitable hardware and or software for identifying and selecting images. For example, tile image identifier and selector may identify and select images as described herein in connection with FIGS. 3 and 5. [0017] Hardware 100 may also include an endpoint 118 and a data image repository 120. Endpoint 118 and endpoint components 128 may be any suitable hardware and/or software for implementing a video conferencing endpoint. Data image repository 120 and 130 may be any suitable storage mechanism for storing data images. For example, data image repository 120 and 130 may include a mass storage device (such as a disk drive, an optical drive, magnetic media, memory, etc.) and database logic (such as database hardware and/or software). In some embodiments, data image repository 120 and/or any other suitable storage device can include a recorded copy of a video conference of which a bit stream may be a part.

[0018] In some embodiments, various components (such as video encoders 106 and 108, video decoders 114 and 124, image identifier and selectors 116 and 126, endpoint 1 18, endpoint components 128, and data image repositories 120 and 130) of hardware 100 can be implemented in one or more general purpose devices such as a computer or a special purpose device such as a client, a server, etc. Any of these general or special purpose devices can include any suitable components such as a processor (which can be a microprocessor, digital signal processor, a controller, etc.), memory, communication interfaces, display controllers, input devices, etc., and can be configured to operate in response to software instructions consistent with the functionality described herein.

[0019] FIG. 2 illustrates an example of a process 200 that can be performed by network appliance 1 12 and/or endpoint 122 in some embodiments. As shown, after process 200 begins at 202, the process receives a video bit stream at 204. The video bit stream can be received in any suitable manner, such as through a transmission over network 110. Next, at 206, the video bit stream can be decoded into video images. Any suitable decoding can be used in some embodiments. Then, sets of data images can be identified in the video images at 208. Any suitable approach to identifying data images can be used in some embodiments. For example, sets of data images can be identified using a process 300 as illustrated in FIG. 3.

[0020] In some embodiments, process 300 can use filters to evaluate video images. For example, filters can be used to determine if a current image is different from a previous image by comparing signatures for the images. Any number of signatures (including none) can be used in a filter, and different filters can use different signatures. A signature can be any suitable representation of an image. For example, in some embodiments, a signature can be derived from coding information available for the image, such as frame type, frame size, macro block (MB) types, etc., and/or pixel-level information for the image, such as average pixel intensity, pixel diversity, etc. In some embodiments, what a signature is based on can be predetermined, can be determined arbitrarily, or can be determined based on content of an image. For example, assuming average pixel intensity for a column is a signature for a certain filter, column selection can be arbitrary (for example, every n-th column) or according to the content in the image (for example, based on pixel diversity in each column).

[0021] Different filters can have different levels of complexity C, in some embodiments, in order to achieve different rates of false positives FP and false negatives FC. False positives are data images that are detected as not matching other data images, but in fact do match. False negatives are data images that do not match other data images, but are detected as matching.

[0022] In some embodiments, filters can be used in a sequential order of passes on an image so that the first filter is the least complex, the next filter is more complex, and subsequent filters are progressively more complex. In this way, if a first filter can be used to determine that a current image is the same as a previous image, then more complex filters do not need to be used.

[0023] FIG. 4 shows a table 400 providing an example of filters for passes 1 through N that can be used to evaluate an image. These filters can have false positive rates FP» through false negative rates FN ₁ through FN*, and complexities C ₁ throUgh C _N , where

[0024] Returning to FIG. 3, after process 300 begins at 302, the process can select a first video image as a current image at 304. Next, process 300 can create a set with the current image and make the current image the previous image at 306. Then, at 308, process 300 can determine if there are any more images, and, if not, end at 310. Otherwise, process 300 can select the next video image as the current image at 312.

[0025] Next, at 314, a first filter can be selected as a current filter. The first filter can be selected based on any suitable criteria or criterion, such as level of complexity, false positive rate, false negative rate, signature used, etc. The current filter can then be applied against the current image to get one or more current signatures at 316. As described above, any suitable type of signature can be obtained.

[0026] Process 300 can next determine if the current image is different from the previous image at 318. Any suitable approach to making mis determination can be used in some embodiments. For example, process 300 can compare the signatures of the current image and the previous image to determine if they are different (versus being identical or similar), [f the current image is determined to be the same as the previous image, then process 300 can add the current image to a set with the previous image and make the current image the previous image at 320 and then branch to 308.

[0027] If the current image is determined to not be the same as the previous image, then process 300 can determine if mere are any more filters to be applied at 322. if so, then process 300 can select the next filter as the current filter at 324 and loop back to 16. Otherwise, process can loop back to 306 to create a set with the current image and make the current image the previous image.

[0028] In some embodiments, animation, sharpening, and other video effects that may be present in data images can be ignored when comparing a current image to a previous image in order to determine whether they should be in the same set. Any suitable approach to determining what animation, sharpening, or other video to ignore, and how to do so, can be used in some embodiments.

10029] Turning back to FIG. 2, after sets of data images have been identified at 208, such as using process 300 of FIG. 3, process 200 can select the best data image in each set of data images at 210. Any suitable approach to determining the best data images can be used in some embodiments. For example, the best data images in each set of data images can be selected using a process 500 as illustrated in FIG. 5 in some embodiments.

[0030] As shown in FIG. 5, after process 500 begins at 502, the process can select the first set of data images at 504. Next, the process can determine, at 506, if the time span across which the images were taken is too short. For example, if a data image appeared for half a second, then the image would not be viewable to a human and thus the image is unlikely to be a valid data image. Any suitable time frame can be used as a basis for determining whether the time span is too short in some embodiments. If it is determined mat the time span was too short, then process 500 can select the next set of data images at 508 and repeat the time span evaluation at 506.

10031] If it is determined at 506 that the time span for the set is not too short, then process 500 can rank the images in the set based on quality at 10. Any suitable mechanism can be used to determine the quality of the images. For example, if a series of images in a set have varying levels of sharpness (e.g., because bandwidth limitations have affected sharpness), then the image with the best sharpness can be selected as having the best quality.

[0032] Next, at 512, process 500 can rank the images in the set based on completeness. Any suitable mechanism can be used to determine the completeness of the images. For example, if data is progressively added to a blank table (for example) in a series of images in a set, the image with the most complete set of data (e.g., the last image) can be selected has having the best completeness.

[0033] The best image in the current set can next be selected, at 514, based on the rankings of quality and completeness. For example, a weighted sum of the rankings can be used to find the best image. Alternatively, in some embodiments, only one of these rankings can be used, or some other factor can be used, to select the best image. In some embodiments, an image can even be selected as the best image arbitrarily, for example, by random (or pseudo random) selection, by being the last image in a set, etc.

[0034] At 516, process 500 can determine if there are any more sets of images, and, if so, select the next set at 508 and loop back to 506. Otherwise, if there are no more sets, then process 500 can end at 518.

[0035] Turning back to FIG. 2, after selecting the best images at 210, process 200 can make the best images available to one or more users at 212. Any suitable approach to doing this can be used in some embodiments. For example, process 200 can store the best images in data image repositories 120 and/or 130. After making the best images available to users at 212, process 200 can wait for the next video bit stream at 214 and men loop back to 204.

]00361 Turning back to FIG. 1 , in some embodiments, endpoim ) 18 and/or 122 (the endpoints) can enable a user to perform certain functions based on the best data images being made available to users. For example, the endpoints (and/or any computer) can enable a user to perform online browsing of data, can allow for late arrivals in a video conference to receive data that was previously shown, can be used to receive data provided in a video conference after the video conference is over, and can allow selection of a portion of a recorded video conference call based on a selection of data that corresponds to it.

[0037] In some embodiments, any suitable computer readable media can be used for storing instructions for performing the processes described herein. For example, in some embodiments, computer readable media can be transitory or non-transitory. For example, non- transitory computer readable media can include media such as magnetic media (such as hard disks, floppy disks, etc.), optical media (such as compact discs, digital video discs, Blu-ray discs, etc.), semiconductor media (such as flash memory, electrically programmable read only memory (EPROM) electrically erasable programmable read only memory (EEPROM), etc.), any suitable media that is not fleering or devoid of any semblance of permanence during transmission, and/or any suitable tangible media. As another example, transitory computer readable media can include signals on networks, in wires, conductors, optical fibers, circuits, any suitable media that is fleeting and devoid of any semblance of permanence during transmission, and/or any suitable intangible media.

[0038] Although the invention has been described and illustrated in the foregoing illustrative embodiments, it is understood that the present disclosure has been made only by way of example, and that numerous changes in the details of implementation of the invention can be made without departing from the spirit and scope of the invention, which is only limited by the claims which follow. Features of the disclosed embodiments can be combined and rearranged in various ways.