Research map / Detection and segmentation / CNN features
PrObeD: Proactive Object Detection Wrapper
Read the paperCheck 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 |
|---|---|---|---|
| add PrObeD wrapper | Faster R-CNN · MS-COCO val | AP: 19.3 -> 31.7 | Helped |
| add PrObeD wrapper | Faster R-CNN + FPN · MS-COCO val | AP: 37.3 -> 38.5 | Helped |
| add PrObeD wrapper | YOLOv5 · MS-COCO val | AP: 48.9 -> 49.4 | Helped |
| add PrObeD wrapper | DeTR · MS-COCO val | AP: 41.9 -> 42.1 | Helped |
| add PrObeD wrapper | DGNet · COD10K | E-measure: 0.833 -> 0.869 | Helped |
| use fixed template instead of image-dependent encoder template (ablation) | PrObeD (Faster R-CNN) · MS-COCO val | AP: 31.7 -> 17.6 | Hurt |
| use one universal learnable template for all images (ablation) | PrObeD (Faster R-CNN) · MS-COCO val | AP: 31.7 -> 19.4 | Hurt |
| remove decoder that recovers template (ablation) | PrObeD (Faster R-CNN) · MS-COCO val | AP: 31.7 -> 25.2 | Hurt |
| add template instead of multiplying (ablation) | PrObeD (Faster R-CNN) · MS-COCO val | AP: 31.7 -> 19.2 | Hurt |
| train Faster R-CNN twice as long without PrObeD (ablation) | Faster R-CNN · MS-COCO val | AP: 19.3 -> 20.1 | Helped |
| train YOLOv5 twice as long without PrObeD (ablation) | YOLOv5 · MS-COCO val | AP: 48.9 -> 48.8 | No clear effect |
| train DGNet twice as long without PrObeD (ablation) | DGNet · COD10K | E-measure: 0.833 -> 0.832 | No clear effect |
Closest papers by meaning
- Sparse DETR: Efficient End-to-End Object Detection with Learnable Sparsity (ICLR 2022 · similarity 0.86)
- Camouflaged Object Detection (CVPR 2020 · similarity 0.86)
- Spatially Invariant Unsupervised Object Detection with Convolutional Neural Networks (AAAI 2019 · similarity 0.85)
- Unsupervised Object Detection with Theoretical Guarantees (NeurIPS 2024 · similarity 0.84)
- MI-DETR: An Object Detection Model with Multi-time Inquiries Mechanism (CVPR 2025 · similarity 0.84)
- DetKDS: Knowledge Distillation Search for Object Detectors (ICML 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.