SeFAR combines dual-level temporal sampling, local temporal reversal, and uncertainty-based loss weighting to set new state-of-the-art results in semi-supervised fine-grained action recognition on FineGym and FineDiving.
InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 26077–26087
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SeFAR: Semi-supervised Fine-grained Action Recognition with Temporal Perturbation and Learning Stabilization
SeFAR combines dual-level temporal sampling, local temporal reversal, and uncertainty-based loss weighting to set new state-of-the-art results in semi-supervised fine-grained action recognition on FineGym and FineDiving.