Research map / Detection and segmentation / CNN features
Learning Detection With Diverse Proposals
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use DPP training and inference instead of Faster R-CNN with NMS | ZF Faster R-CNN · PASCAL VOC2007 test, trained on VOC2007 trainval, IoU 0.5 | mAP: 60.45 -> 62.21 | Helped |
| Use DPP training and inference instead of Faster R-CNN with NMS | ZF Faster R-CNN · PASCAL VOC2007 test, trained on VOC2007 trainval, IoU 0.7 | mAP: 34.6 -> 36.8 | Helped |
| Use DPP training and inference instead of Faster R-CNN with NMS | ZF Faster R-CNN · PASCAL VOC2007 test, trained on VOC2012 trainval, IoU 0.5 | mAP: 60.9 -> 63.0 | Helped |
| Use DPP training and inference instead of Faster R-CNN with NMS | VGG_CNN_M_1024 Faster R-CNN · MS COCO val | average precision, IoU 0.5-0.95: 15.0 -> 15.5 | Helped |
| Use DPP training and inference instead of Faster R-CNN with NMS | VGG_CNN_M_1024 Faster R-CNN · MS COCO val, 100 detections | average recall: 23.6 -> 26.0 | Helped |
| Raise semantic similarity to the fourth power rather than use it to the first power (ablation) | LDDP inference · MS COCO minival | average precision, IoU 0.5: 32.0 -> 32.3 | Helped |
| Remove semantic similarity from the kernel (ablation) | LDDP inference · MS COCO minival | average precision, IoU 0.5-0.95: 15.4 -> 14.5 | Hurt |
| Limit the number of proposals to 100 (ablation) | LDDP · PASCAL VOC2007 test | mAP: 62.2% -> 60.4% | Hurt |
Closest papers by meaning
- Proposal-Contrastive Pretraining for Object Detection from Fewer Data (ICLR 2023 · similarity 0.87)
- Group R-CNN for Weakly Semi-Supervised Object Detection With Points (CVPR 2022 · similarity 0.87)
- Spatially Invariant Unsupervised Object Detection with Convolutional Neural Networks (AAAI 2019 · similarity 0.86)
- You Reap What You Sow: Using Videos to Generate High Precision Object Proposals for Weakly-Supervised Object Detection (CVPR 2019 · similarity 0.86)
- Hybrid Proposal Refiner: Revisiting DETR Series from the Faster R-CNN Perspective (CVPR 2024 · similarity 0.86)
- DetKDS: Knowledge Distillation Search for Object Detectors (ICML 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.