Abstract
Due to the numerous potential applications in visual surveillance and nighttime driving, recognizing human action in low-light conditions remains a difficult problem in computer vision. Existing methods separate action recognition and dark enhancement into two distinct steps to accomplish this task. However, Isolating the recognition and enhancement impedes end-to-end learning of the space-time representation for video action classification. This paper presents a domain adaptation-based action recognition approach that uses adversarial learning in cross-domain settings to learn cross-domain action recognition. Supervised learning can train it on a large amount of labelled data from the source domain (daytime action sequences). However, it uses deep domain invariant features to perform unsupervised learning on many unlabelled data from the target domain (night-time action sequences). The resulting augmented model, named 3D-DiNet can be trained using standard backpropagation with an additional layer. It achieves SOTA performance on InFAR and XD145 actions datasets.
Original language | English |
---|---|
Title of host publication | 2022 International Conference on Digital Image Computing |
Subtitle of host publication | Techniques and Applications (DICTA 2022) |
Place of Publication | United States |
Publisher | IEEE Xplore |
Pages | 279-285 |
Number of pages | 7 |
ISBN (Electronic) | 9781665456425 |
ISBN (Print) | 9781665456432 |
DOIs | |
Publication status | E-pub ahead of print - 10 Feb 2023 |
Event | International Conference on Digital Image Computing: Techniques and Applications (DICTA 2022) - Rydges World Square, Sydney, Australia Duration: 30 Nov 2022 → 02 Dec 2022 http://dicta2022.dictaconference.org/ (Conference website) http://dicta2022.dictaconference.org/full_program.html#oral4 (Program) https://ieeexplore.ieee.org/xpl/conhome/10034558/proceeding (Conference proceedings) http://dicta2022.dictaconference.org/DICTA%202022%20CFP.pdf (Call for papers) |
Publication series
Name | 2022 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2022 |
---|
Conference
Conference | International Conference on Digital Image Computing |
---|---|
Country/Territory | Australia |
City | Sydney |
Period | 30/11/22 → 02/12/22 |
Other | The International Conference on Digital Image Computing: Techniques and Applications (DICTA) is the main Australian Conference on computer vision, image processing, pattern recognition, and related areas. DICTA was established in 1991 as the premier conference of the Australian Pattern Recognition Society (APRS). |
Internet address |
|