</> Code with Papers by Syntology

Datasets › HACS

HACS (Human Action Clips and Segments)

Introduced by Hang Zhao et al. in HACS: Human Action Clips and Segments Dataset for Recognition and Temporal Localization archive 2025-07-28

HACS is a dataset for human action recognition. It uses a taxonomy of 200 action classes, which is identical to that of the ActivityNet-v1.3 dataset. It has 504K videos retrieved from YouTube. Each one is strictly shorter than 4 minutes, and the average length is 2.6 minutes. A total of 1.5M clips of 2-second duration are sparsely sampled by methods based on both uniform randomness and consensus/disagreement of image classifiers. 0.6M and 0.9M clips are annotated as positive and negative samples, respectively.

Authors split the collection into training, validation and testing sets of size 1.4M, 50K and 50K clips, which are sampled from 492K, 6K and 6K videos, respectively.

HomepageBenchmarksPapers JSON

Benchmarks archive 2025-07-28

All 2 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Temporal Action Localization HACS RDFA-S6 (InternVideo2-6B) Average-mAP 45.8 Enhancing Temporal Action Localization: Advanced S6... lsy0882/RDFA-S6 12 Compare
Action Recognition HACS InternVideo2-6B Top 1 Accuracy 97.0 InternVideo2: Scaling Foundation Models for Multimodal... opengvlab/internvideo +1 8 Compare

Papers archive 2025-07-28

12 shown of 12 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 75. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
Enhancing Temporal Action Localization: Advanced S6 Modeling with Recurrent Mechanism 1 1 18 Jul 2024 not harvested
DyFADet: Dynamic Feature Aggregation for Temporal Action Detection 1 1 3 Jul 2024 ran 6 of 12 samples (6 unverified; 12 pointer-only for licence)
InternVideo2: Scaling Foundation Models for Multimodal Video Understanding 2 3 22 Mar 2024 not harvested
Video Mamba Suite: State Space Model as a Versatile Alternative for Video Understanding 1 1 14 Mar 2024 ran 8 of 13 samples (5 unverified)
Temporal Action Localization with Enhanced Instant Discriminability 3 1 11 Sep 2023 not harvested
TriDet: Temporal Action Detection with Relative Boundary Modeling 1 2 13 Mar 2023 ran 5 of 15 samples (10 unverified)
InternVideo: General Video Foundation Models via Generative and Discriminative Learning 2 1 6 Dec 2022 ran 3 of 3 samples (0 unverified)
UniFormerV2: Spatiotemporal Learning by Arming Image ViTs with Video UniFormer 2 1 22 Sep 2022 not harvested
Low-Fidelity Video Encoder Optimization for Temporal Action Localization 0 1 1 Dec 2021 not harvested
End-to-end Temporal Action Detection with Transformer 1 1 18 Jun 2021 ran 2 of 3 samples (1 unverified)
Learn to cycle: Time-consistent feature discovery for action recognition 1 6 15 Jun 2020 not harvested
HACS: Human Action Clips and Segments Dataset for Recognition and Temporal Localization 2 1 26 Dec 2017 ran 0 of 1 samples (1 unverified)

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

Action RecognitionAction LocalizationTemporal Action Localization

License archive 2025-07-28

BSD 3-Clause

Modalities archive 2025-07-28

Videos

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

  • HACS

1 variant name, as the archive lists them.

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