Research map / Video / Action recognition
MGSampler: An Explainable Sampling Strategy for Video Action Recognition
Yuan Zhi, Zhan Tong, Limin Wang, Gangshan Wu
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 |
|---|---|---|---|
| replace TSN uniform sampling with motion-guided sampling (feature difference) | TSM ResNet50, 8 frames · Something-Something V2 | top-1 accuracy: 57.9 -> 60.1 | Improved |
| replace TSN uniform sampling with motion-guided sampling (feature difference) | TSM ResNet50, 8 frames · Something-Something V1 | top-1 accuracy: 45.6 -> 47.3 | Improved |
| replace TSN sampling with motion-guided sampling (feature difference) | TSM ResNet50, 8 frames · Diving48 | top-1 accuracy: 35.2 -> 37.4 | Improved |
| replace TSN sampling with motion-guided sampling (feature difference) | TSM ResNet50, 8 frames · UCF101 | top-1 accuracy: 94.5 -> 95.2 | Improved |
| replace TSN sampling with motion-guided sampling (feature difference) | TSM ResNet50, 8 frames · HMDB51 | top-1 accuracy: 72.6 -> 73.8 | Improved |
| replace TSN sampling with motion-guided sampling (feature difference) | TSM ResNet50, 8 frames · Jester | top-1 accuracy: 96.5 -> 97.5 | Improved |
| use feature-level difference instead of image-level difference as motion representation (ablation) | MGSampler · Something-Something V2 | top-1 accuracy: 59.8 -> 60.1 | Improved |
| set smoothing mu=0.5 instead of mu=1 (original motion magnitude) (ablation) | MGSampler image diff · Something-Something V1 | top-1 accuracy: 45.7 -> 47.1 | Improved |
| set smoothing mu=2 instead of mu=0.5 (exaggerate motion differences) (ablation) | MGSampler image diff · Something-Something V1 | top-1 accuracy: 47.1 -> 42.8 | Hurt |
| use cumulative-motion-distribution sampling instead of motion-magnitude-only sampling (ablation) | MGSampler · Something-Something V1 | top-1 accuracy: 41.5 -> 47.3 | Improved |
| use motion-guided sampling instead of fixed stride sampling | TSM ResNet50 · Something-Something V2 | top-1 accuracy: 53.4 -> 60.1 | Improved |
| add MGSampler to TSM ResNet101 with 8 frames | TSM · Something-Something V1 | top-1 accuracy: 46.9 -> 47.8 | Improved |
Closest papers by meaning
- SCSampler: Sampling Salient Clips From Video for Efficient Action Recognition (ICCV 2019 · similarity 0.91)
- No Frame Left Behind: Full Video Action Recognition (CVPR 2021 · similarity 0.89)
- Few-Shot Action Recognition via Intra- and Inter-Video Information Maximization (TPAMI 2026 · similarity 0.88)
- Learn2Augment: Learning to Composite Videos for Data Augmentation in Action Recognition (ECCV 2022 · similarity 0.88)
- Taylor Videos for Action Recognition (ICML 2024 · similarity 0.87)
- ActionAtlas: A VideoQA Benchmark for Domain-specialized Action Recognition (NeurIPS 2024 · similarity 0.86)
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.
