Research map / Video / Action recognition
BMN: Boundary-Matching Network for Temporal Action Proposal Generation
Tianwei Lin, Xiao Liu, Xin Li, Errui Ding, Shilei Wen
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 PEM trained separately to TEM (ablation) | BMN · ActivityNet-1.3 val | AUC: 65.17 -> 66.43 | Improved |
| add PEM trained separately to TEM (ablation) | BSN · ActivityNet-1.3 val | AUC: 64.80 -> 66.17 | Improved |
| train TEM and PEM jointly end-to-end (ablation) | BMN · ActivityNet-1.3 val | AR@100: 74.36 -> 75.01 | Improved |
| train TEM and PEM jointly end-to-end (ablation) | BMN · ActivityNet-1.3 val | AUC: 66.43 -> 67.10 | Improved |
| train TEM and PEM jointly end-to-end (ablation) | BMN · ActivityNet-1.3 val | total inference time (s): 0.069 -> 0.052 | Improved |
| replace BSN proposal-feature scoring with boundary-matching confidence map | BSN · ActivityNet-1.3 val | total inference time (s): 0.629 -> 0.069 | Improved |
| replace BSN with BMN (TEM only) (ablation) | BSN TEM · ActivityNet-1.3 val | AR@100: 73.57 -> 73.72 | No clear effect |
| replace BSN with BMN (end-to-end) | BSN · ActivityNet-1.3 val | AUC: 66.17 -> 67.10 | Improved |
| use Soft-NMS instead of Greedy-NMS (ablation) | BMN · THUMOS-14 (C3D features) | AR@50: 29.04 -> 32.73 | Improved |
| use Soft-NMS instead of Greedy-NMS (ablation) | BMN · THUMOS-14 (C3D features) | AR@1000: 60.96 -> 60.44 | Hurt |
| train on seen+unseen categories instead of seen only (ablation) | BMN · ActivityNet-1.3 unseen subset | AUC: 64.47 -> 65.05 | No clear effect |
| use C3D to two-stream features | BMN+SNMS · THUMOS-14 | AR@100: 40.68 -> 47.72 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CTAP — CTAP: Complementary Temporal Action Proposal Generation (ECCV 2018) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- BSN: Boundary Sensitive Network for Temporal Action Proposal Generation (ECCV 2018 · similarity 0.94)
- BSN++: Complementary Boundary Regressor with Scale-Balanced Relation Modeling for Temporal Action Proposal Generation (AAAI 2021 · similarity 0.93)
- DCAN: Improving Temporal Action Detection via Dual Context Aggregation (AAAI 2022 · similarity 0.92)
- Efficient Temporal Action Segmentation via Boundary-aware Query Voting (NeurIPS 2024 · similarity 0.86)
- HyrCap: Hybrid Rank-Calibration of Action Proposals for Temporal Event Understanding (NeurIPS 2026 · similarity 0.86)
- Realigning Confidence with Temporal Saliency Information for Point-Level Weakly-Supervised Temporal Action Localization (CVPR 2024 · 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.
