Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Fine-grained Action Recognition research benchmarks · SOTA2 Research

Fine-grained Action Recognition

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Representation Learning via Global Temporal Alignment and Cycle-Consistency

46.54Accuracy

12

Feb 26, 2026

Representation Learning via Global Temporal Alignment and Cycle-Consistency

49.51Accuracy

12

Feb 26, 2026

Gym99-skeleton severe temporal corruption (Sev.)

89.1Top-1 Acc

9

Feb 26, 2026

Gym99-skeleton (moderate temporal corruption (Mod.))

90.6Top-1 Acc

9

Feb 26, 2026

Gym99-skeleton minor temporal corruption (Min.)

92.1Top-1 Acc

9

Feb 26, 2026

Gym288-skeleton (severe temporal corruption)

78.1Top-1 Acc

9

Feb 26, 2026

Gym288-skeleton moderate temporal corruption (Mod.)

79.7Top-1 Acc

9

Feb 26, 2026

Gym288-skeleton minor temporal corruption (Min.)

81.5Top-1 Acc

9

Feb 26, 2026

FineGym 288 (original clips)

24.16Accuracy

6

Feb 26, 2026

FineGym99 (original clips)

27.81Accuracy

6

Feb 26, 2026

99.79Accuracy (Armstand)

4

Feb 26, 2026

Popular datasets

Gym FineGym MTL-AQA

Follow for updates

Join SOTA2 Contact