Research map / Detection and segmentation / Weakly supervised detection and segmentation
C-MIL: Continuation Multiple Instance Learning for Weakly Supervised Object Detection
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 |
|---|---|---|---|
| Replace MIL with continuation instance selection and detector estimation | MIL on PASCAL VOC 2007 with VGGF · PASCAL VOC 2007 | mAP: 36.0 vs. 40.7 | Helped |
| Use continuation instance selection only (ablation) | C-MIL on PASCAL VOC 2007 with VGGF · PASCAL VOC 2007 | mAP: 36.0 vs. 39.0 | Helped |
| Use continuation object estimation only (ablation) | C-MIL on PASCAL VOC 2007 with VGGF · PASCAL VOC 2007 | mAP: 36.0 vs. 37.4 | Helped |
| Combine continuation instance selection and object estimation (ablation) | C-MIL on PASCAL VOC 2007 with VGGF · PASCAL VOC 2007 | mAP: 40.7% vs. 36.0% | Helped |
| Use the Log continuation function rather than the other tested schedules (ablation) | C-MIL on PASCAL VOC 2007 with VGGF · PASCAL VOC 2007 | mAP: 37.9, 37.6, 38.3, 37.1 -> 40.7 | Helped |
| Use the Log continuation function rather than the other tested schedules (ablation) | C-MIL on PASCAL VOC 2007 with VGGF · PASCAL VOC 2007 | CorLoc: 58.9, 57.4, 58.4, 56.4 -> 59.5 | Helped |
| Apply C-MIL instead of MELM | Weakly supervised detection with VGGF · PASCAL VOC 2007 test | mAP: 40.7% vs. 38.4% | Helped |
| Apply C-MIL instead of MELM | Weakly supervised detection with VGG16 · PASCAL VOC 2007 test | mAP: 50.5% vs. 47.3% | Helped |
| Apply C-MIL instead of MELM | Weakly supervised detection with VGG16 · PASCAL VOC 2012 | mAP: 46.7% vs. 42.4% | Helped |
| Apply C-MIL instead of WeakRPN | Weakly supervised localization with VGG16 · PASCAL VOC 2012 | CorLoc: 67.4% vs. 64.9% | Helped |
| Apply C-MIL instead of WeakRPN | Object localization with VGG16 · PASCAL VOC 2007 trainval | mAP: 65.0% vs. 63.8% | Helped |
Closest papers by meaning
- Object Instance Mining for Weakly Supervised Object Detection (AAAI 2020 · similarity 0.90)
- Instance Mining with Class Feature Banks for Weakly Supervised Object Detection (AAAI 2021 · similarity 0.90)
- Multiple Instance Detection Network With Online Instance Classifier Refinement (CVPR 2017 · similarity 0.90)
- Instance-Aware, Context-Focused, and Memory-Efficient Weakly Supervised Object Detection (CVPR 2020 · similarity 0.89)
- Weakly Supervised Few-Shot Object Detection with DETR (AAAI 2024 · similarity 0.87)
- Weakly Supervised Open-Vocabulary Object Detection (AAAI 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.