Title:
LIGHT WEIGHT MULTI-BRANCH AND MULTI-SCALE PERSON RE-IDENTIFICATION
Document Type and Number:
WIPO Patent Application WO/2021/120157
Kind Code:
A1
Abstract:
A system for lightweight multi-branch and multi-scale (LMBMS) re-identification is described herein. The system includes a convolutional neural network trained for person identification, wherein the convolutional neural network comprises a series of residual blocks that obtain input from a head network of the convolutional neural network. The system also includes a plurality of refine blocks, wherein one or more refine blocks take as input features from a residual block of the series of residual blocks, wherein the features are at input at different scales and different resolutions and an output of the plurality of refine blocks is a plurality of features in a same feature space. A channel-wise attention mechanism may merge the plurality of features and generate final dynamic features.
Inventors:
TAO HONGZHI (CN)
LI QIANG (CN)
LI QIANG (CN)
Application Number:
PCT/CN2019/126906
Publication Date:
June 24, 2021
Filing Date:
December 20, 2019
Export Citation:
Assignee:
INTEL CORP (US)
TAO HONGZHI (CN)
LI QIANG (CN)
TAO HONGZHI (CN)
LI QIANG (CN)
International Classes:
G06K9/00; G06K9/62; G06N3/04
Foreign References:
| CN109886090A | 2019-06-14 | |||
| US10176405B1 | 2019-01-08 | |||
| CN109948587A | 2019-06-28 | |||
| CN110516670A | 2019-11-29 | |||
| US8615105B1 | 2013-12-24 |
Other References:
WANG YICHENG; CHEN ZHENZHONG; WU FENG; WANG GANG: "Person Re-identification with Cascaded Pairwise Convolutions", 2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, IEEE, 18 June 2018 (2018-06-18), pages 1470 - 1478, XP033476110, DOI: 10.1109/CVPR.2018.00159
Attorney, Agent or Firm:
CHINA PATENT AGENT (H.K.) LTD. (CN)
Download PDF: