Research map / Video / Action recognition
Local Temporal Bilinear Pooling for Fine-Grained Action Parsing
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 |
|---|---|---|---|
| Replace max pooling with decoupled local temporal bilinear pooling | TCED · 50 Salads-mid | accuracy: 64.7 -> 66.3 | Helped |
| Replace max pooling with decoupled local temporal bilinear pooling | TCED · 50 Salads-mid | edit score: 59.8 -> 62.5 | Helped |
| Replace max pooling with decoupled local temporal bilinear pooling | TCED · 50 Salads-mid | F1 score: 68.0 -> 68.9 | Helped |
| Replace max pooling with decoupled local temporal bilinear pooling | TCED · 50 Salads-eval | edit score: 72.2 -> 71.3 | Hurt |
| Replace max pooling with decoupled local temporal bilinear pooling | TCED · 50 Salads-eval | accuracy: 73.4 -> 75.9 | Helped |
| Replace max pooling with decoupled local temporal bilinear pooling | TCED · GTEA | F1 score: 75.2 -> 76.8 | Helped |
| Replace max pooling with decoupled local temporal bilinear pooling | TCED · JIGSAWS suturing | F1 score: 90.3 -> 91.4 | Helped |
| Replace averaging with learnable local filter weights (ablation) | Coupled and decoupled bilinear pooling · 50 Salads-mid | evaluation metrics | Helped |
| Use decoupled rather than coupled bilinear pooling (ablation) | TCED on 50 Salads-mid · 50 Salads-mid | accuracy: 64.73 -> 66.3 | Helped |
| Combine first- and second-order components rather than use either component alone (ablation) | Decoupled bilinear pooling · 50 Salads-mid | evaluation metrics | Helped |
| Use NReLU rather than ReLU with l2 normalization (ablation) | Bilinear pooling on 50 Salads-mid · 50 Salads-mid | F1 score: 68.04 -> 68.93 | Mixed |
| Replace explicit bilinear features with the exact low-dimensional representation phi_d (ablation) | TCED on 50 Salads-mid · 50 Salads-mid | F1 score: 68.9 -> 69.0 | Helped |
Closest papers by meaning
- Temporal Convolutional Networks for Action Segmentation and Detection (CVPR 2017 · similarity 0.89)
- Temporal Bilinear Networks for Video Action Recognition (AAAI 2019 · similarity 0.87)
- Learning End-to-end Video Classification with Rank-Pooling (ICML 2016 · similarity 0.87)
- Weakly Supervised Action Localization by Sparse Temporal Pooling Network (CVPR 2018 · similarity 0.87)
- Dual DETRs for Multi-Label Temporal Action Detection (CVPR 2024 · similarity 0.86)
- SeFAR: Semi-supervised Fine-grained Action Recognition with Temporal Perturbation and Learning Stabilization (AAAI 2025 · 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.