Research map / Detection and segmentation / CNN features
PTSEFormer: Progressive Temporal-Spatial Enhanced TransFormer towards Video Object Detection
Han Wang, Jun Tang, Xiaodong Liu, Shanyan Guan, Rong Xie, Li Song
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 STAM to single frame baseline (ablation) | PTSEFormer · ImageNet VID val (ResNet-50) | mAP: 81.2 -> 84.5 | Improved |
| add TFAM on top of STAM (ablation) | PTSEFormer · ImageNet VID val (ResNet-50) | mAP: 84.5 -> 87.4 | Improved |
| add Query Assembling Model instead of DETR object queries (ablation) | PTSEFormer · ImageNet VID val (ResNet-50) | mAP: 86.1 -> 87.4 | Improved |
| add Gated Correlation instead of original transformer decoder (ablation) | PTSEFormer · ImageNet VID val (ResNet-50) | mAP: 86.7 -> 87.4 | Improved |
| add Residual Gated Correlation (ablation) | PTSEFormer · ImageNet VID val (ResNet-50) | mAP: 86.3 -> 87.4 | Improved |
| add multi-scale features (ablation) | PTSEFormer · ImageNet VID val (ResNet-50) | mAP: 86.4 -> 87.4 | Improved |
| add Seq-NMS post-processing | MEGA · ImageNet VID val (ResNet-101) | mAP: 82.9 -> 84.5 | Improved |
| add Seq-NMS post-processing | SELSA · ImageNet VID val (ResNet-101) | mAP: 80.3 -> 82.5 | Improved |
| use ResNet-101 instead of ResNet-50 | PTSEFormer · ImageNet VID val | mAP: 87.4 -> 88.1 | Improved |
| use PTSEFormer instead of TransVOD (no post-processing) | Deformable DETR VOD · ImageNet VID val (ResNet-101) | mAP: 81.9 -> 88.1 | Improved |
Later papers that compared with this method
1 later accepted paper reports PTSEFormer as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Practical Video Object Detection via Feature Selection and Aggregation (IJCV 2026 · better on 1 of 1 shared result)
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.
- TransVOD — TransVOD: End-to-End Video Object Detection With Spatial-Temporal Transformers (TPAMI 2022) · 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
- Context Enhanced Transformer for Single Image Object Detection in Video Data (AAAI 2024 · similarity 0.89)
- TransVOD: End-to-End Video Object Detection With Spatial-Temporal Transformers (TPAMI 2022 · similarity 0.87)
- TGBFormer: Transformer-GraphFormer Blender Network for Video Object Detection (AAAI 2025 · similarity 0.85)
- Knowing Your Target: Target-Aware Transformer Makes Better Spatio-Temporal Video Grounding (ICLR 2025 · similarity 0.84)
- QDETRv: Query-Guided DETR for One-Shot Object Localization in Videos (AAAI 2024 · similarity 0.84)
- EGSST: Event-based Graph Spatiotemporal Sensitive Transformer for Object Detection (NeurIPS 2024 · similarity 0.84)
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.
