CVFeb 25, 2023

Temporal Segment Transformer for Action Segmentation

arXiv:2302.13074v15 citationsh-index: 60
Originality Incremental advance
AI Analysis

This work addresses the challenge of long-range temporal modeling in video activity understanding for computer vision applications, representing an incremental improvement over existing predict-and-refine methods.

The paper tackled the problem of noisy and inaccurate segment representations in action segmentation from untrimmed videos by proposing a temporal segment transformer that uses attention for joint segment relation modeling and denoising, achieving state-of-the-art accuracy on benchmarks like 50Salads, GTEA, and Breakfast.

Recognizing human actions from untrimmed videos is an important task in activity understanding, and poses unique challenges in modeling long-range temporal relations. Recent works adopt a predict-and-refine strategy which converts an initial prediction to action segments for global context modeling. However, the generated segment representations are often noisy and exhibit inaccurate segment boundaries, over-segmentation and other problems. To deal with these issues, we propose an attention based approach which we call \textit{temporal segment transformer}, for joint segment relation modeling and denoising. The main idea is to denoise segment representations using attention between segment and frame representations, and also use inter-segment attention to capture temporal correlations between segments. The refined segment representations are used to predict action labels and adjust segment boundaries, and a final action segmentation is produced based on voting from segment masks. We show that this novel architecture achieves state-of-the-art accuracy on the popular 50Salads, GTEA and Breakfast benchmarks. We also conduct extensive ablations to demonstrate the effectiveness of different components of our design.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes