Research map / Video / Action recognition
Gate-Shift-Fuse for Video Action Recognition
Swathikiran Sudhakaran, Sérgio Escalera, Oswald Lanz
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 learnable channel fusion in place of sum fusion (ablation) | GSF (BNInception) · Something-Something V1 (BNInception, 8 frames) | accuracy: 47.24 -> 48.44 | Improved |
| make spatial gating learnable instead of fixed to 1 (TSM-like) (ablation) | GSF (BNInception) · Something-Something V1 (BNInception, 8 frames) | accuracy: 45.57 -> 48.44 | Improved |
| make gating learnable instead of fixed to -1 (differencing) (ablation) | GSF (BNInception) · Something-Something V1 (BNInception, 8 frames) | accuracy: 45.89 -> 48.44 | Improved |
| add gating (non-zero) over fixed gating 0 (TSN) (ablation) | GSF (BNInception) · Something-Something V1 (BNInception, 8 frames) | accuracy: 17.25 -> 48.44 | Improved |
| add GSF to all 10 Inception blocks instead of last block only (ablation) | TSN BNInception · Something-Something V1 (BNInception, 8 frames) | accuracy: 31.9 -> 48.44 | Improved |
| add GSF to 5 blocks instead of 1 (ablation) | TSN BNInception · Something-Something V1 (BNInception, 8 frames) | accuracy: 31.9 -> 45.24 | Improved |
| plug GSF into the least-convolution Inception branch rather than all branches (ablation) | GSF (BNInception) · Something-Something V1 (BNInception, 8 frames) | accuracy: 46.55 -> 48.44 | Improved |
| shift all channels instead of 12.5% (ablation) | GSF (BNInception) · Something-Something V1 (BNInception, 8 frames) | accuracy: 45.62 -> 48.44 | Improved |
| shift only 25% of channels instead of 100% (ablation) | GSF (ResNet-50) · Something-Something V1 (ResNet-50, 8 frames) | accuracy: 47.63 -> 50.17 | Improved |
| add GSF to 16 bottleneck layers instead of 4 (ablation) | TSN ResNet-50 · Something-Something V1 (ResNet-50, 8 frames) | accuracy: 45.97 -> 50.17 | Improved |
| apply GSF after conv2 vs after conv3 of ResNet bottleneck (ablation) | GSF (ResNet-50) · Something-Something V1 (ResNet-50, 8 frames) | accuracy: 47.15 -> 47.63 | Mixed |
| add MotionSqueeze module to GSF ResNet-50 (16 frames, 3x2 clips) | GSF ResNet-50 · Something-Something V1 | accuracy: 52.33 -> 54.04 | Improved |
Closest papers by meaning
- Gate-Shift Networks for Video Action Recognition (CVPR 2020 · similarity 0.93)
- Deep Analysis of CNN-Based Spatio-Temporal Representations for Action Recognition (CVPR 2021 · similarity 0.87)
- Depthwise Spatio-Temporal STFT Convolutional Neural Networks for Human Action Recognition (TPAMI 2021 · similarity 0.87)
- Spatiotemporal Fusion in 3D CNNs: A Probabilistic View (CVPR 2020 · similarity 0.87)
- A Closer Look at Spatiotemporal Convolutions for Action Recognition (CVPR 2018 · similarity 0.87)
- Spatiotemporal Pyramid Network for Video Action Recognition (CVPR 2017 · 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.
