Research map / Detection and segmentation / CNN features
Fast Convergence of DETR With Spatially Modulated Co-Attention
Peng Gao, Minghang Zheng, Xiaogang Wang, Jifeng Dai, Hongsheng Li
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 head-shared spatially modulated co-attention (ablation) | DETR-R50 · COCO 2017 val (50 epochs) | mAP: 34.8 -> 40.2 | Improved |
| switch to multi-head spatial modulation from head-shared (ablation) | SMCA · COCO 2017 val (50 epochs) | mAP: 40.2 -> 41.0 | Improved |
| predict the Gaussian center instead of no modulation (fixed scale) (ablation) | SMCA · COCO 2017 val (50 epochs) | mAP: 34.8 -> 38.5 | Improved |
| predict a single scale instead of fixed scale (ablation) | SMCA · COCO 2017 val (50 epochs) | mAP: 38.5 -> 40.4 | Improved |
| predict independent height and width scales instead of single scale (ablation) | SMCA · COCO 2017 val (50 epochs) | mAP: 40.4 -> 41.0 | Improved |
| change batch size from 8 to 32 for head-shared modulation (ablation) | SMCA · COCO 2017 val (50 epochs) | mAP: 40.2 -> 39.9 | No clear effect |
| add multi-scale feature encoding with 2Intra-Multi-2Intra self-attention (ablation) | SMCA · COCO 2017 val (50 epochs) | mAP: 41.0 -> 43.7 | Improved |
| add scale-selection attention (ablation) | SMCA multi-scale · COCO 2017 val (50 epochs) | mAP: 42.6 -> 43.7 | Improved |
| use 2Intra-Multi-2Intra instead of 5 intra-scale layers (ablation) | SMCA multi-scale · COCO 2017 val (50 epochs) | mAP: 43.3 -> 43.7 | Improved |
| share weights of intra-scale self-attention FFN instead of no sharing (ablation) | SMCA multi-scale · COCO 2017 val (50 epochs) | mAP: 42.3 -> 43.0 | Improved |
| train SMCA longer from 50 to 108 epochs | SMCA R50 · COCO 2017 val | mAP: 43.7 -> 45.6 | Improved |
| use SMCA instead of DETR with 10x fewer epochs | SMCA R50 · COCO 2017 val | mAP: 42.0 -> 45.6 | Improved |
Closest papers by meaning
- Sparse DETR: Efficient End-to-End Object Detection with Learnable Sparsity (ICLR 2022 · similarity 0.89)
- CF-DETR: Coarse-to-Fine Transformers for End-to-End Object Detection (AAAI 2022 · similarity 0.89)
- DAC-DETR: Divide the Attention Layers and Conquer (NeurIPS 2023 · similarity 0.89)
- CP-DETR: Concept Prompt Guide DETR Toward Stronger Universal Object Detection (AAAI 2025 · similarity 0.87)
- Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement (CVPR 2024 · similarity 0.87)
- Hybrid Proposal Refiner: Revisiting DETR Series from the Faster R-CNN Perspective (CVPR 2024 · similarity 0.87)
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.
