Abstract
One of the major challenges in person Re-Identification (ReID) is the inconsistent visual appearance of a person. Current works on visual feature and distance metric learning have achieved significant achievements, but still suffer from the limited robustness to pose variations, viewpoint changes, etc. This makes person ReID among multiple cameras still challenging. This work is motivated to learn mid-level human attributes which are robust to visual appearance variations and could be used as efficient features for person matching. We propose a supervised multi-task learning framework which considers attribute label information with joint identification-verification network to simultaneously learn an attribute-semantic and identity-discriminative feature representation. Specifically, this framework adopts the part-based deep neural network and learn three different tasks simultaneously: person identification, person verifications and attribute identification, so as to discover and capture concurrently complementary discriminative information about a person image from global and local image features and mid-level attribute features in one deep neural network. With the multi-task learning architecture, we obtain a discriminative model that reaches a synergy in distinguishing different person images, as manifested with the competitive accuracy on three person ReID datasets: Market1501, DukeMTMC-reID and VIPeR.
Highlights
Person re-identification (ReID) aims to identify a query person by finding the same persons among a set of gallery images or videos
The convolutional neural networks (CNNs) based approaches for ReID can be broadly classified into two categories: identification mode and verification mode
Our contributions are three-fold: (1) we propose a new multi-task deep learning architecture which considers mid-level semantic attribute information by taking advantage of both multi-class identification and binary verification supervision signal to simultaneously learn an attribute-semantic and identity-discriminative feature representation more effectively; (2) we systematically analyze the relationship between person attribute and identity labels, and further exploit the interaction of verification and identification supervision; (3) we compare our approach with several state-of-art methods on three popular ReId datasets: Market1501 [27], DukeMTMC-reID [28] and VIPeR [29]
Summary
Person re-identification (ReID) aims to identify a query person by finding the same persons among a set of gallery images or videos. The task is challenging owing to the dramatic changes in visual appearance from different camera views with non-overlapping visual fields resulting from variations in postures, view angles, occlusion, illumination, low resolution and background clutter [7]. Traditional researches on this topic mainly focus on two separate phases independently: either carefully feature designing The identification mode utilizes a classification loss function (e.g. cross entropy loss) to learn a mapping function from raw images to person identity directly This mode could make full use of the label information of the dataset, but it is prone to overfitting due to the limited
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
