@inproceedings{c5c5f276a29a4510a6e3b7c271e2d992,
title = "Flow Dynamics Correction for Action Recognition",
abstract = "Various research studies indicate that action recognition performance highly depends on the types of motions being extracted and how accurate the human actions are represented. In this paper, we investigate different optical flow, and features extracted from these optical flow that capturing both short-term and long-term motion dynamics. We perform power normalization on the magnitude component of optical flow for flow dynamics correction to boost subtle or dampen sudden motions. We show that existing action recognition models which rely on optical flow are able to get performance boosted with our corrected optical flow. To further improve performance, we integrate our corrected flow dynamics into popular models through a simple hallucination step by selecting only the best performing optical flow features, and we show that by 'translating' the CNN feature maps into these optical flow features with different scales of motions leads to the new state-of-the-art performance on several benchmarks including HMDB-51, YUP++, fine-grained action recognition on MPII Cooking Activities, and large-scale Charades.",
keywords = "action recognition, flow correction, hallucination, optical flow, power normalization",
author = "Lei Wang and Piotr Koniusz",
note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 49th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 ; Conference date: 14-04-2024 Through 19-04-2024",
year = "2024",
doi = "10.1109/ICASSP48485.2024.10446223",
language = "English",
isbn = "979-8-3503-4486-8",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "3795--3799",
booktitle = "2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024 - Proceedings",
address = "United States",
}