Skip to Content
No preview available


Download Analytics Citations

Export to: EndNote  |  Zotero  |  Mendeley


This file is not currently in any collections.

Re-ID-AR: Improved Person Re-identification in Video via Joint Weakly Supervised Action Recognition [dataset] Open Access

We uniquely consider the task of joint person re-identification (Re-ID) and action recognition in video as a multi-task problem. In addition to the broader potential of joint Re-ID and action recognition within the context of automated multi-camera surveillance, we show that the consideration of action recognition in addition to Re-ID results in a model that learns discriminative feature representations that both improve Re-ID performance and are capable of providing viable per-view (clip-wise) action recognition. Our approach uses a single 2D Convolutional Neural Network (CNN) architecture comprising a common ResNet50-IBN backbone CNN architecture, to extract frame-level features with subsequent temporal attention for clip-level feature extraction, followed by two subbranches:- the IDentification (sub-)Network (IDN) for person Re-ID and the ActionRecognition (sub-)Network for per-view action recognition. The IDN comprises a single fully connected layer while the ARN comprises multiple attention blocks on a one-to-one ratio with the number of actions to be recognised. This is subsequently trained as a jointRe-ID and action recognition task using a combination of two task-specific, multi-loss terms via weakly labelled actions obtained over two leading benchmark Re-ID datasets (MARS & LPW). Our consideration of Re-ID and action recognition as a multi-task problem results in a multi-branch 2D CNN architecture that outperforms prior work in the field (rank-1 (mAP) – MARS: 93.21%(87.23%), LPW: 79.60%) without any reliance 3D convolutions or multi-stream networks architectures as found in other contemporary work. Our work represents the first benchmark performance for such a joint Re-ID and action recognition video understanding task, hitherto unapproached in the literature, and is accompanied by a new public dataset of supplementary action labels for the seminal MARS and LPW Re-ID datasets.


Resource type
Creator: Alsehaim, Aishah 1
Data collector: Alsehaim, Aishah 1
Breckon, Toby 1
1 Durham University, UK
Research methods
Other description
Action recognition
Convolutional Neural Networks
Computer vision
Neural networks (Computer science)--Mathematical models
Cited in
Creative Commons Attribution 4.0 International (CC BY)

Durham University
Date Created

File Details

A.A. Alsehaim
Date Uploaded
Date Modified
11 November 2021, 12:11:38
Audit Status
Audits have not yet been run on this file.
File format: x-rar (RAR archive data, v8a, os: MS-DOS, Rar)
Mime type: application/x-rar
File size: 7159368
Filename: RE-ID-AR.rar
Original checksum: b88c253b97690d6edd8e5f34878139b3
Activity of users you follow
User Activity Date
User N. Syrotiuk has transferred Re-ID-AR: Improved Person Re-identification in Video via Joint Weakly Supervised Action Recognition [dataset] to user A.A. Alsehaim 9 months ago