Papers › FACT: Frame-Action Cross-Attention Temporal Modeling for Efficient Action Segmentation

FACT: Frame-Action Cross-Attention Temporal Modeling for Efficient Action Segmentation

1 Jan 2024CVPR 2024 1archive 2025-07-28

Zijia Lu, Ehsan Elhamifar

We study supervised action segmentation whose goal is to predict framewise action labels of a video. To capture temporal dependencies over long horizons prior works either improve framewise features with transformer or refine framewise predictions with learned action features. However they are computationally costly and ignore that frame and action features contain complimentary information which can be leveraged to enhance both features and improve temporal modeling. Therefore we propose an efficient Frame-Action Cross-attention Temporal modeling (FACT) framework that performs temporal modeling with frame and action features in parallel and leverage this parallelism to achieve iterative bidirectional information transfer between the features and refine them. FACT network contains (i) a frame branch to learn frame-level information with convolutions and frame features (ii) an action branch to learn action-level dependencies with transformers and action tokens and (iii) cross-attentions to allow communication between the two branches. We also propose a new matching loss to ensure each action token uniquely encodes an action segment thus better captures its semantics. Thanks to our architecture we can also leverage textual transcripts of videos to help action segmentation. We evaluate FACT on four video datasets (two egocentric and two third-person) for action segmentation with and without transcripts showing that it significantly improves the state-of-the-art accuracy while enjoys lower computational cost (3 times faster) than existing transformer-based methods

PaperPDFCode

Code

zijialewislu/cvpr2024-fact officialmentioned in paperpytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Action SegmentationSegmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Segmentation Breakfast FACT (efficient hybrid of convolution and transformer model) Acc 76.2 #2 of 37 Archive leaderboard report
Action Segmentation Breakfast FACT (efficient hybrid of convolution and transformer model) Average F1 74.7 #2 of 37 Archive leaderboard report
Action Segmentation Breakfast FACT (efficient hybrid of convolution and transformer model) Edit 79.7 #2 of 37 Archive leaderboard report
Action Segmentation Breakfast FACT (efficient hybrid of convolution and transformer model) F1@10% 81.4 #2 of 37 Archive leaderboard report
Action Segmentation Breakfast FACT (efficient hybrid of convolution and transformer model) F1@25% 76.5 #2 of 37 Archive leaderboard report
Action Segmentation Breakfast FACT (efficient hybrid of convolution and transformer model) F1@50% 66.2 #2 of 37 Archive leaderboard report
Action Segmentation GTEA FACT Acc 84.5 #2 of 28 Archive leaderboard report
Action Segmentation GTEA FACT Edit 93.5 #2 of 28 Archive leaderboard report
Action Segmentation GTEA FACT F1@10% 96.1 #2 of 28 Archive leaderboard report
Action Segmentation GTEA FACT F1@25% 95.6 #2 of 28 Archive leaderboard report
Action Segmentation GTEA FACT F1@50% 87.5 #2 of 28 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections