SSA2D: Single Shot Actor-Action Detection in Videos (Student Abstract)
Abstract
We propose a single-shot approach for actor-action detection in videos. The existing approaches use a two-step process, which rely on Region Proposal Network (RPN), where the action is estimated based on the detected proposals followed by post-processing such as non-maximal suppression. While effective in terms of performance, these methods pose limitations in scalability for dense video scenes with a high memory requirement for thousand of proposals, which leads to slow processing time. We propose SSA2D, a unified end-to-end deep network, which performs joint actor-action detection in a single-shot without the need of any proposals and post-processing, making it memory as well as time efficient.
Cite
Text
Rana and Rawat. "SSA2D: Single Shot Actor-Action Detection in Videos (Student Abstract)." AAAI Conference on Artificial Intelligence, 2021. doi:10.1609/AAAI.V35I18.17934Markdown
[Rana and Rawat. "SSA2D: Single Shot Actor-Action Detection in Videos (Student Abstract)." AAAI Conference on Artificial Intelligence, 2021.](https://mlanthology.org/aaai/2021/rana2021aaai-ssa/) doi:10.1609/AAAI.V35I18.17934BibTeX
@inproceedings{rana2021aaai-ssa,
title = {{SSA2D: Single Shot Actor-Action Detection in Videos (Student Abstract)}},
author = {Rana, Aayush Jung and Rawat, Yogesh S.},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2021},
pages = {15875-15876},
doi = {10.1609/AAAI.V35I18.17934},
url = {https://mlanthology.org/aaai/2021/rana2021aaai-ssa/}
}