Research map / Video / Action recognition
Temporal Deformable Residual Networks for Action Segmentation in Videos
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 |
|---|---|---|---|
| Use deformable rather than standard temporal convolutions (ablation) | TDRN · 50Salads (mid) | F1@{10,25,50}: 70.2,65.4,56.3 -> 72.9,68.5,57.2 | Helped |
| Use deformable rather than standard temporal convolutions (ablation) | TDRN · GTEA | F1@{10,25,50}: 77.4,71.3,59.1 -> 79.2,74.4,62.7 | Helped |
| Use deformable rather than standard temporal convolutions (ablation) | TDRN · JIGSAWS | F1@{10}: 91.4 -> 92.9 | Helped |
| Add TUnet-like connections (ablation) | TDRN · 50Salads (mid) | F1@{10,25,50}: 72.9,68.5,57.2 -> 69.6,65.0,53.6 | Hurt |
| Add TUnet-like connections (ablation) | TDRN · GTEA | F1@{10,25,50}: 79.2,74.4,62.7 -> 78.1,73.8,62.2 | Hurt |
| Add TUnet-like connections (ablation) | TDRN · JIGSAWS | F1@{10}: 92.9 -> 92.1 | Hurt |
| Increase temporal kernel size (ablation) | TDRN · 50Salads (mid) | F1@10 | Mixed |
| Vary the number of DTRMs (ablation) | TDRN · 50Salads (mid) | F1@10 | Mixed |
Closest papers by meaning
- Temporal Convolutional Networks for Action Segmentation and Detection (CVPR 2017 · similarity 0.90)
- MS-TCN: Multi-Stage Temporal Convolutional Network for Action Segmentation (CVPR 2019 · similarity 0.89)
- TDN: Temporal Difference Networks for Efficient Action Recognition (CVPR 2021 · similarity 0.89)
- Adaptive Residual Quantization for Memory-Efficient Temporal Action Segmentation (NeurIPS 2026 · similarity 0.88)
- FACT: Frame-Action Cross-Attention Temporal Modeling for Efficient Action Segmentation (CVPR 2024 · similarity 0.86)
- OnlineTAS: An Online Baseline for Temporal Action Segmentation (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.