End-to-End Spatio-Temporal Action Localisation with Video Transformers

Alexey A. Gritsenko, Xuehan Xiong, Josip Djolonga, Mostafa Dehghani, Chen Sun, Mario Lucic, Cordelia Schmid, Anurag Arnab

Keywords: action localisation, video understanding

TL;DR: We develop a performant, end-to-end spatio-temporal action localisation model which directly predicts tubelets, even when weaker frame-level ground-truth annotations are used.

Abstract: The most performant spatio-temporal action localisation models use external person proposals and complex external memory banks. We propose a fully end-to-end, transformer based model that directly ingests an input video, and outputs tubelets -- a sequence of bounding boxes and the action classes at each frame. Our flexible model can be trained with either sparse bounding-box supervision on individual frames, or full tubelet annotations. And in both cases, it predicts coherent tubelets as the output. Moreover, our end-to-end model requires no additional pre-processing in the form of proposals, or post-processing in terms of non-maximal suppression. We perform extensive ablation experiments, and significantly advance the state-of-the-art results on four different spatio-temporal action localisation benchmarks with both sparse keyframes and full tubelet annotations.

Primary Area: representation learning for computer vision, audio, language, and other modalities

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 2440