Research map / Detection and segmentation / CNN features
Enriched Feature Guided Refinement Network for Object Detection
Jing Nie, Rao Muhammad Anwer, Hisham Cholakkal, Fahad Shahbaz Khan, Yanwei Pang, Ling Shao
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 feature enrichment scheme (ablation) | SSD VGG16 320 · PASCAL VOC 2007 | mAP: 77.2 -> 79.4 | Improved |
| add cascaded refinement scheme (ablation) | SSD VGG16 320 · PASCAL VOC 2007 | mAP: 77.2 -> 81.0 | Improved |
| add feature enrichment scheme (ablation) | SSD VGG16 320 · MS COCO minival | AP: 24.4 -> 29.1 | Improved |
| add cascaded refinement scheme (ablation) | SSD VGG16 320 · MS COCO minival | AP: 24.4 -> 31.1 | Improved |
| combine both schemes (full EFGRNet) (ablation) | SSD VGG16 320 · MS COCO minival | AP: 24.4 -> 33.0 | Improved |
| combine both schemes (full EFGRNet) (ablation) | SSD VGG16 320 · PASCAL VOC 2007 | mAP: 77.2 -> 81.4 | Improved |
| add second dilated branch to MSCF (r=1,2) (ablation) | MSCF · PASCAL VOC 2007 | mAP: 78.7 -> 79.0 | Improved |
| add third dilated branch (r=4) to MSCF (ablation) | MSCF · PASCAL VOC 2007 | mAP: 79.0 -> 79.4 | Improved |
| add more branches with other dilation rates beyond three (ablation) | MSCF · PASCAL VOC 2007 | mAP | No clear effect |
| use standard deformable convolution offsets instead of dilated convolution (ablation) | FGRM · PASCAL VOC 2007 | mAP: 80.2 -> 80.5 | Improved |
| generate deformable offsets from B1x position offsets (dx,dy) instead of feature-based offsets (ablation) | FGRM · PASCAL VOC 2007 | mAP: 80.5 -> 81.0 | Improved |
| use only scale offsets (dh,dw) from B1x for deformable offsets (ablation) | FGRM · PASCAL VOC 2007 | mAP: 81.0 -> 80.3 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- CoupleNet — CoupleNet: Coupling Global Structure With Local Parts for Object Detection (ICCV 2017) · better on 0 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
- Single-Shot Refinement Neural Network for Object Detection (CVPR 2018 · similarity 0.90)
- Dynamic Refinement Network for Oriented and Densely Packed Object Detection (CVPR 2020 · similarity 0.89)
- Learning Rich Features at High-Speed for Single-Shot Object Detection (ICCV 2019 · similarity 0.89)
- Context Refinement for Object Detection (ECCV 2018 · similarity 0.89)
- Single-Shot Object Detection With Enriched Semantics (CVPR 2018 · similarity 0.88)
- Long-tailed Object Detection Pretraining: Dynamic Rebalancing Contrastive Learning with Dual Reconstruction (NeurIPS 2024 · similarity 0.86)
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.
