Research map / Video / Action recognition
Semantic Adversarial Network with Multi-Scale Pyramid Attention for Video Classification
Read the paperCheck 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 |
|---|---|---|---|
| Add MPA to the spatial baseline (ablation) | Spatial Network · HMDB51 split 1 | accuracy: 52.0% -> 54.0% | Helped |
| Add MPA to the spatial baseline (ablation) | Spatial Network · UCF101 split 1 | accuracy: 84.5% -> 85.2% | Helped |
| Add SAL to the MPA model (ablation) | Spatial Network · HMDB51 split 1 | accuracy: 54.0% -> 54.4% | Helped |
| Add SAL to the MPA model (ablation) | Spatial Network · UCF101 split 1 | accuracy: 85.2% -> 86.1% | Helped |
| Use the proposed RGB spatial branch instead of TSN spatial | Video classification · HMDB51 split 1 | accuracy: 52.0% -> 54.4% | Helped |
| Use the proposed differential-image temporal branch instead of TSN temporal | Video classification · HMDB51 split 1 | accuracy: 57.1% -> 57.8% | Helped |
| Use the proposed combined streams instead of TSN | Video classification · HMDB51 split 1 | accuracy: 62.5% -> 63.7% | Helped |
| Use the proposed RGB spatial branch instead of TSN spatial | Video classification · UCF101 split 1 | accuracy: 84.5% -> 86.1% | Helped |
| Use the proposed differential-image temporal branch instead of TSN temporal | Video classification · UCF101 split 1 | accuracy: 85.4% -> 86.8% | Helped |
| Use the proposed combined streams instead of TSN | Video classification · UCF101 split 1 | accuracy: 89.9% -> 91.2% | Helped |
Closest papers by meaning
- Convolutional Two-Stream Network Fusion for Video Action Recognition (CVPR 2016 · similarity 0.87)
- AssembleNet: Searching for Multi-Stream Neural Connectivity in Video Architectures (ICLR 2020 · similarity 0.86)
- Spatiotemporal Pyramid Network for Video Action Recognition (CVPR 2017 · similarity 0.86)
- Harnessing Object and Scene Semantics for Large-Scale Video Understanding (CVPR 2016 · similarity 0.86)
- SmallBigNet: Integrating Core and Contextual Views for Video Classification (CVPR 2020 · similarity 0.85)
- A Dual Attention Network with Semantic Embedding for Few-Shot Learning (AAAI 2019 · similarity 0.85)
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.