Papers › ActionFormer: Localizing Moments of Actions with Transformers

ActionFormer: Localizing Moments of Actions with Transformers

16 Feb 2022arXiv:2202.07925archive 2025-07-28

Chenlin Zhang, Jianxin Wu, Yin Li

Self-attention based Transformer models have demonstrated impressive results for image classification and object detection, and more recently for video understanding. Inspired by this success, we investigate the application of Transformer networks for temporal action localization in videos. To this end, we present ActionFormer -- a simple yet powerful model to identify actions in time and recognize their categories in a single shot, without using action proposals or relying on pre-defined anchor windows. ActionFormer combines a multiscale feature representation with local self-attention, and uses a light-weighted decoder to classify every moment in time and estimate the corresponding action boundaries. We show that this orchestrated design results in major improvements upon prior works. Without bells and whistles, ActionFormer achieves 71.0% mAP at tIoU=0.5 on THUMOS14, outperforming the best prior model by 14.1 absolute percentage points. Further, ActionFormer demonstrates strong results on ActivityNet 1.3 (36.6% average mAP) and EPIC-Kitchens 100 (+13.5% average mAP over prior works). Our code is available at http://github.com/happyharrycn/actionformer_release.

PaperPDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

happyharrycn/actionformer_release officialmentioned in papermentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to SnAG: Scalable and Accurate Video Grounding (arXiv:2404.02257; that paper's own run record: 13 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 9 where Syntology's instrument failed) · 7 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

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

Tasks

Action LocalizationAction RecognitionDecoderTemporal Action LocalizationVideo Understandingaudio-visual event localizationimage-classificationobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Temporal Action Localization ActivityNet-1.3 ActionFormer (TSP feautures) mAP 36.6 #15 of 33 Archive leaderboard report
Temporal Action Localization ActivityNet-1.3 ActionFormer (TSP feautures) mAP IOU@0.5 54.7 #15 of 33 Archive leaderboard report
Temporal Action Localization ActivityNet-1.3 ActionFormer (TSP feautures) mAP IOU@0.75 37.8 #15 of 33 Archive leaderboard report
Temporal Action Localization ActivityNet-1.3 ActionFormer (TSP feautures) mAP IOU@0.95 8.4 #15 of 33 Archive leaderboard report
Temporal Action Localization EPIC-KITCHENS-100 ActionFormer (verb) Avg mAP (0.1-0.5) 23.5 #4 of 6 Archive leaderboard report
Temporal Action Localization EPIC-KITCHENS-100 ActionFormer (verb) mAP IOU@0.1 26.6 #4 of 6 Archive leaderboard report
Temporal Action Localization EPIC-KITCHENS-100 ActionFormer (verb) mAP IOU@0.2 25.4 #4 of 6 Archive leaderboard report
Temporal Action Localization EPIC-KITCHENS-100 ActionFormer (verb) mAP IOU@0.3 24.2 #4 of 6 Archive leaderboard report
Temporal Action Localization EPIC-KITCHENS-100 ActionFormer (verb) mAP IOU@0.4 22.3 #4 of 6 Archive leaderboard report
Temporal Action Localization EPIC-KITCHENS-100 ActionFormer (verb) mAP IOU@0.5 19.1 #4 of 6 Archive leaderboard report
Temporal Action Localization THUMOS’14 ActionFormer (I3D features) Avg mAP (0.3:0.7) 66.8 #13 of 42 Archive leaderboard report
Temporal Action Localization THUMOS’14 ActionFormer (I3D features) mAP IOU@0.3 82.1 #13 of 42 Archive leaderboard report
Temporal Action Localization THUMOS’14 ActionFormer (I3D features) mAP IOU@0.4 77.8 #13 of 42 Archive leaderboard report
Temporal Action Localization THUMOS’14 ActionFormer (I3D features) mAP IOU@0.5 71.0 #13 of 42 Archive leaderboard report
Temporal Action Localization THUMOS’14 ActionFormer (I3D features) mAP IOU@0.6 59.4 #13 of 42 Archive leaderboard report
Temporal Action Localization THUMOS’14 ActionFormer (I3D features) mAP IOU@0.7 43.9 #13 of 42 Archive leaderboard report
audio-visual event localization UnAV-100 ActionFormer mAP 42.2 #2 of 2 Archive leaderboard report
audio-visual event localization UnAV-100 ActionFormer AP@IOU0.5 43.5 #2 of 2 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.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

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