Papers › Learning Action Completeness from Points for Weakly-supervised Temporal Action Localization

Learning Action Completeness from Points for Weakly-supervised Temporal Action Localization

11 Aug 2021ICCV 2021 10arXiv:2108.05029archive 2025-07-28

Pilhyeon Lee, Hyeran Byun

We tackle the problem of localizing temporal intervals of actions with only a single frame label for each action instance for training. Owing to label sparsity, existing work fails to learn action completeness, resulting in fragmentary action predictions. In this paper, we propose a novel framework, where dense pseudo-labels are generated to provide completeness guidance for the model. Concretely, we first select pseudo background points to supplement point-level action labels. Then, by taking the points as seeds, we search for the optimal sequence that is likely to contain complete action instances while agreeing with the seeds. To learn completeness from the obtained sequence, we introduce two novel losses that contrast action instances with background ones in terms of action score and feature similarity, respectively. Experimental results demonstrate that our completeness guidance indeed helps the model to locate complete action instances, leading to large performance gains especially under high IoU thresholds. Moreover, we demonstrate the superiority of our method over existing state-of-the-art methods on four benchmarks: THUMOS'14, GTEA, BEOID, and ActivityNet. Notably, our method even performs comparably to recent fully-supervised methods, at the 6 times cheaper annotation cost. Our code is available at https://github.com/Pilhyeon.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Pilhyeon/Learning-Action-Completeness-from-Points officialmentioned on GitHubpytorch 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 LocalizationTemporal Action LocalizationWeakly Supervised Action LocalizationWeakly-supervised Temporal Action Localization

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Weakly Supervised Action Localization ActivityNet-1.2 LACP Mean mAP 26.8 #5 of 19 Archive leaderboard report
Weakly Supervised Action Localization ActivityNet-1.2 LACP mAP@0.5 44 #5 of 19 Archive leaderboard report
Weakly Supervised Action Localization ActivityNet-1.3 LACP mAP@0.5 40.4 #9 of 17 Archive leaderboard report
Weakly Supervised Action Localization ActivityNet-1.3 LACP mAP@0.5:0.95 25.1 #9 of 17 Archive leaderboard report
Weakly Supervised Action Localization BEOID LACP mAP@0.1:0.7 51.8 #2 of 5 Archive leaderboard report
Weakly Supervised Action Localization BEOID LACP mAP@0.5 42.7 #2 of 5 Archive leaderboard report
Weakly Supervised Action Localization GTEA LACP mAP@0.1:0.7 43.5 #3 of 6 Archive leaderboard report
Weakly Supervised Action Localization GTEA LACP mAP@0.5 33.9 #3 of 6 Archive leaderboard report
Weakly Supervised Action Localization THUMOS 2014 LACP mAP@0.1:0.5 62.7 #2 of 30 Archive leaderboard report
Weakly Supervised Action Localization THUMOS 2014 LACP mAP@0.1:0.7 52.8 #2 of 30 Archive leaderboard report
Weakly Supervised Action Localization THUMOS 2014 LACP mAP@0.5 45.3 #2 of 30 Archive leaderboard report
Weakly Supervised Action Localization THUMOS14 LACP avg-mAP (0.1-0.5) 62.7 #2 of 12 Archive leaderboard report
Weakly Supervised Action Localization THUMOS14 LACP avg-mAP (0.1:0.7) 52.8 #2 of 12 Archive leaderboard report
Weakly Supervised Action Localization THUMOS14 LACP avg-mAP (0.3-0.7) 44.5 #2 of 12 Archive leaderboard report
Weakly Supervised Action Localization THUMOS’14 LACP mAP@0.5 45.3 #1 of 13 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