Papers › Boundary-Aware Cascade Networks for Temporal Action Segmentation

Boundary-Aware Cascade Networks for Temporal Action Segmentation

1 Aug 2020ECCV 2020 8archive 2025-07-28

Zhenzhi Wang, Ziteng Gao, Li-Min Wang, Zhifeng Li, Gangshan Wu

Identifying human action segments in an untrimmed video is still challenging due to boundary ambiguity and over-segmentation issues. To address these problems, we present a new boundary-aware cascade network by introducing two novel components. First, we devise a new cascading paradigm, called Stage Cascade, to enable our model to have adaptive receptive fields and more confident predictions for ambiguous frames. Second, we design a general and principled smoothing operation, termed as local barrier pooling, to aggregate local predictions by leveraging semantic boundary information. Moreover, these two components can be jointly fine-tuned in an end-to-end manner. We perform experiments on three challenging datasets: 50Salads, GTEA and Breakfast dataset, demonstrating that our framework significantly out-performs the current state-of-the-art methods. The code is available at https://github.com/MCG-NJU/BCN.

PaperPDFCode

Code

MCG-NJU/BCN officialmentioned in paperpytorch 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 SegmentationTemporal Action Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Segmentation 50 Salads BCN Acc 84.4 #17 of 28 Archive leaderboard report
Action Segmentation 50 Salads BCN Edit 74.3 #17 of 28 Archive leaderboard report
Action Segmentation 50 Salads BCN F1@10% 82.3 #17 of 28 Archive leaderboard report
Action Segmentation 50 Salads BCN F1@25% 81.3 #17 of 28 Archive leaderboard report
Action Segmentation 50 Salads BCN F1@50% 74 #17 of 28 Archive leaderboard report
Action Segmentation Breakfast BCN Acc 70.4 #23 of 37 Archive leaderboard report
Action Segmentation Breakfast BCN Average F1 63.1 #23 of 37 Archive leaderboard report
Action Segmentation Breakfast BCN Edit 66.2 #23 of 37 Archive leaderboard report
Action Segmentation Breakfast BCN F1@10% 68.7 #23 of 37 Archive leaderboard report
Action Segmentation Breakfast BCN F1@25% 65.5 #23 of 37 Archive leaderboard report
Action Segmentation Breakfast BCN F1@50% 55.0 #23 of 37 Archive leaderboard report
Action Segmentation GTEA BCN Acc 79.8 #16 of 28 Archive leaderboard report
Action Segmentation GTEA BCN Edit 84.4 #16 of 28 Archive leaderboard report
Action Segmentation GTEA BCN F1@10% 88.5 #16 of 28 Archive leaderboard report
Action Segmentation GTEA BCN F1@25% 87.1 #16 of 28 Archive leaderboard report
Action Segmentation GTEA BCN F1@50% 77.3 #16 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