Research map / Video / Action recognition
Video Imprint Segmentation for Temporal Action Detection in Untrimmed Videos
Zhanning Gao, Le Wang, Qilin Zhang, Zhenxing Niu, Nanning Zheng, Gang Hua
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 |
|---|---|---|---|
| increase imprint grid from 24x24 to 48x48 (ablation) | DBS · THUMOS'14 | mAP@0.5: 27.8 -> 31.5 | Improved |
| increase imprint grid from 48x48 to 64x64 (ablation) | DBS · THUMOS'14 | mAP@0.5: 31.5 -> 31.7 | No clear effect |
| use 3 conv layers instead of 1 (ablation) | DBS FCN · THUMOS'14 | mAP@0.5: 27.4 -> 31.5 | Improved |
| use 4 conv layers instead of 3 (ablation) | DBS FCN · THUMOS'14 | mAP@0.5: 31.5 -> 30.7 | Hurt |
| 3x3 kernels instead of 1x1 kernels (ablation) | DBS FCN · THUMOS'14 | mAP@0.5: 29.6 -> 31.5 | Improved |
| add cropped ground-truth action instances to training (ablation) | DBS · THUMOS'14 | mAP@0.5: 24.6 -> 28.2 | Improved |
| add UCF101 trimmed videos to training (ablation) | DBS · THUMOS'14 | mAP@0.5: 24.6 -> 27.1 | Improved |
| combine both augmentations with original data (ablation) | DBS · THUMOS'14 | mAP@0.5: 24.6 -> 31.5 | Improved |
| fuse RGB and flow predictions instead of RGB only (ablation) | DBS · THUMOS'14 | mAP@0.5: 31.5 -> 34.3 | Improved |
| flow feature instead of RGB feature (ablation) | DBS · THUMOS'14 | mAP@0.5: 31.5 -> 31.2 | No clear effect |
| DBS vs CDC for per-frame labeling | DBS · THUMOS'14 per-frame | mAP: 44.4 -> 47.8 | Improved |
| DBS vs best baselines on detection across IoU | DBS · THUMOS'14 | mAP: 31.0 -> 34.3 at IoU 0.5; 51.9 -> 50.6 at IoU 0.3 | Mixed |
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.
- R-C3D — R-C3D: Region Convolutional 3D Network for Temporal Activity Detection (ICCV 2017) · better on 1 of 1 shared result
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
- Deep Learning-Based Action Detection in Untrimmed Videos: A Survey (TPAMI 2022 · similarity 0.90)
- ASM-Loc: Action-Aware Segment Modeling for Weakly-Supervised Temporal Action Localization (CVPR 2022 · similarity 0.89)
- Iterative Contrast-Classify for Semi-supervised Temporal Action Segmentation (AAAI 2022 · similarity 0.88)
- Temporally-Weighted Hierarchical Clustering for Unsupervised Action Segmentation (CVPR 2021 · similarity 0.88)
- Fast and Unsupervised Action Boundary Detection for Action Segmentation (CVPR 2022 · similarity 0.88)
- Unsupervised Action Segmentation via Fast Learning of Semantically Consistent Actoms (AAAI 2024 · similarity 0.88)
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.
