Research map / Video / Action recognition
SF-Net: Single-Frame Supervision for Temporal Action Localization
Fan Ma, Linchao Zhu, Yi Yang, Shengxin Zha, Gourab Kundu, Matt Feiszli, Zheng Shou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add single-frame supervision on top of video-level labels | weak baseline · GTEA | mAP AVG 0.1-0.7 (%): 7.0 -> 30.5 | Improved |
| add single-frame supervision on top of video-level labels | weak baseline · BEOID | mAP AVG 0.1-0.7 (%): 8.7 -> 19.7 | Improved |
| add single-frame supervision on top of video-level labels | weak baseline · THUMOS14 | mAP AVG 0.1-0.7 (%): 30.8 -> 31.7 | Improved |
| add pseudo background frame mining (ablation) | SF · GTEA | mAP AVG 0.1-0.7 (%): 30.5 -> 28.0 | Hurt |
| add pseudo background frame mining (ablation) | SF · BEOID | mAP AVG 0.1-0.7 (%): 19.7 -> 21.7 | Improved |
| add pseudo background frame mining (ablation) | SF · THUMOS14 | mAP AVG 0.1-0.7 (%): 31.7 -> 33.9 | Improved |
| add actionness module (ablation) | SFB · GTEA | mAP AVG 0.1-0.7 (%): 28.0 -> 26.4 | Hurt |
| add actionness module (ablation) | SFB · BEOID | mAP AVG 0.1-0.7 (%): 21.7 -> 27.1 | Improved |
| add actionness module (ablation) | SFB · THUMOS14 | mAP AVG 0.1-0.7 (%): 33.9 -> 39.9 | Improved |
| add action frame mining (expansion) (ablation) | SFBA · GTEA | mAP AVG 0.1-0.7 (%): 26.4 -> 31.0 | Improved |
| add action frame mining (expansion) (ablation) | SFBA · BEOID | mAP AVG 0.1-0.7 (%): 27.1 -> 30.1 | Improved |
| add action frame mining (expansion) (ablation) | SFBA · THUMOS14 | mAP AVG 0.1-0.7 (%): 39.9 -> 40.6 | Improved |
Closest papers by meaning
- PivoTAL: Prior-Driven Supervision for Weakly-Supervised Temporal Action Localization (CVPR 2023 · similarity 0.92)
- Background-Click Supervision for Temporal Action Localization (TPAMI 2021 · similarity 0.92)
- Pointly-Supervised Action Localization (IJCV 2018 · similarity 0.92)
- Bridge the Gap: From Weak to Full Supervision for Temporal Action Localization with PseudoFormer (CVPR 2025 · similarity 0.91)
- Weakly-Supervised Temporal Action Localization by Inferring Salient Snippet-Feature (AAAI 2024 · similarity 0.91)
- Realigning Confidence with Temporal Saliency Information for Point-Level Weakly-Supervised Temporal Action Localization (CVPR 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
