Research map / Video / Action recognition
Collaborative Spatiotemporal Feature Learning for Video Action Recognition
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use collaborative three-view CoST instead of C3D | ResNet-50 action recognition · Moments in Time validation | Top-1 accuracy (%): 29.0 -> 30.1 | Helped |
| Use collaborative three-view CoST instead of C3D | ResNet-101 action recognition · Kinetics validation | Top-1 accuracy (%): 74.5 -> 75.5 | Helped |
| Predict view coefficients from each sample rather than learn fixed coefficients (ablation) | CoST with ResNet-50 · Moments in Time validation | Top-1 accuracy (%): 29.3 -> 30.1 | Helped |
| Predict view coefficients from each sample rather than learn fixed coefficients (ablation) | CoST with ResNet-50 · Kinetics validation | Top-1 accuracy (%): 73.6 -> 74.1 | Helped |
| Add shared convolution weights across the three views (ablation) | CoST with ResNet-50 · Moments in Time validation | Top-1 accuracy (%): 29.0 -> 30.1 | Helped |
| Add shared convolution weights across the three views (ablation) | CoST with ResNet-50 · Kinetics validation | Top-1 accuracy (%): 73.2 -> 74.1 | Helped |
| Increase input from 8 to 32 frames | CoST with ResNet-101 · Moments in Time validation | Top-1 accuracy (%): 31.5 -> 32.4 | Helped |
| Increase input from 8 to 32 frames | CoST with ResNet-101 · Kinetics validation | Top-1 accuracy (%): 75.5 -> 77.5 | Helped |
Closest papers by meaning
- A Closer Look at Spatiotemporal Convolutions for Action Recognition (CVPR 2018 · similarity 0.91)
- Spatiotemporal Pyramid Network for Video Action Recognition (CVPR 2017 · similarity 0.90)
- Spatiotemporal Residual Networks for Video Action Recognition (NeurIPS 2016 · similarity 0.90)
- Spatiotemporal Fusion in 3D CNNs: A Probabilistic View (CVPR 2020 · similarity 0.90)
- Deep Analysis of CNN-Based Spatio-Temporal Representations for Action Recognition (CVPR 2021 · similarity 0.89)
- Building a Multi-modal Spatiotemporal Expert for Zero-shot Action Recognition with CLIP (AAAI 2025 · similarity 0.87)
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.