Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly Supervised Cascaded Convolutional Networks
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 |
|---|---|---|---|
| add the weakly supervised segmentation stage to the two-stage cascade (ablation) | WCCN with VGG16 · PASCAL VOC 2007 test | detection mAP: 41.4 -> 42.8 | Helped |
| add the weakly supervised segmentation stage to the two-stage cascade (ablation) | WCCN with VGG16 · PASCAL VOC 2007 test | classification mAP: 89.9 -> 90.9 | Helped |
| add the weakly supervised segmentation stage to the two-stage cascade (ablation) | WCCN with VGG16 · PASCAL VOC 2007 trainval positive set | CorLoc: 54.6 -> 56.7 | Helped |
| add the weakly supervised segmentation stage to the two-stage cascade (ablation) | WCCN with VGG16 · PASCAL VOC 2010 test | detection mAP: 37.8 -> 39.5 | Helped |
| add the weakly supervised segmentation stage to the two-stage cascade (ablation) | WCCN with VGG16 · PASCAL VOC 2012 test | detection mAP: 36.4 -> 37.9 | Helped |
| add the weakly supervised segmentation stage to the two-stage cascade (ablation) | WCCN with VGG16 · ILSVRC 2013 validation | detection mAP: 14.6 -> 16.3 | Helped |
| replace VGG16 with AlexNet (ablation) | WCCN three-stage cascade · PASCAL VOC 2007 test | detection mAP: 42.8 -> 37.3 | Hurt |
| replace global average pooling with global max pooling (ablation) | WCCN location network · PASCAL VOC experiments | performance | Hurt |
| use EdgeBox proposals instead of SelectiveSearch (ablation) | WCCN two-stage AlexNet detector · PASCAL VOC 2007 test | detection performance: approximately 1.5% | Helped |
| train Fast R-CNN using WCCN-generated pseudo ground-truth boxes instead of the full WCCN detection pipeline | PASCAL VOC 2007 test · PASCAL VOC 2007 test | average precision: 0.3% | Helped |
Closest papers by meaning
- Weakly Supervised Deep Detection Networks (CVPR 2016 · similarity 0.93)
- Multi-Evidence Filtering and Fusion for Multi-Label Classification, Object Detection and Semantic Segmentation Based on Weakly Supervised Learning (CVPR 2018 · similarity 0.88)
- Multiple Instance Detection Network With Online Instance Classifier Refinement (CVPR 2017 · similarity 0.88)
- CRAFT Objects From Images (CVPR 2016 · similarity 0.87)
- Cascade R-CNN: Delving Into High Quality Object Detection (CVPR 2018 · similarity 0.87)
- Instance-Aware, Context-Focused, and Memory-Efficient Weakly Supervised Object Detection (CVPR 2020 · 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.