Research map / Detection and segmentation / Open-set and few-shot detection
Weak-Shot Object Detection Through Mutual Knowledge Transfer
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 Knowledge Transfer (KT) loss (ablation) | MIL module with objectness+entropy transfer · VOC 2007 test, COCO-60 source | mAP: +0.7 (60.9 to 61.6) | Helped |
| add Consistency Filtering (CF-g) to pseudo-label refinement (ablation) | PG module refinement · VOC 2007 test, COCO-60 source | mAP: +1.2 (60.9 to 62.1) | Helped |
| add both KT loss and CF (ablation) | WSHOD mutual transfer framework · VOC 2007 test, COCO-60 source | mAP: +2.0 (60.9 to 62.9) | Helped |
| use generative filter criterion (CF-g) instead of discriminative (CF-d) (ablation) | CF method · VOC 2007 test, COCO-60 source | mAP: 62.9 vs 60.5 | Helped |
| omit random noise injection in CF (direct threshold filtering) (ablation) | CF method · VOC 2007 test, COCO-60 source | mAP: 61.9 vs 62.9 | Hurt |
| use continuous noise regions vs scattered regions in CF (ablation) | CF method · VOC 2007 test, COCO-60 source | mAP: 62.9 vs 61.6 (zeros noise) | Helped |
| proposed method vs previous best WSHOD (TraMaS), single scale, no distillation | WSHOD on VOC 2007 with COCO-60 source · VOC 2007 test | mAP: +2.0 (62.9 vs 60.9) | Helped |
| proposed method vs TraMaS on ILSVRC-179 source | WSHOD on VOC 2007 with ILSVRC-179 source · VOC 2007 test, ILSVRC-179 source | mAP: +2.1 (60.4 vs 58.3) | Helped |
Closest papers by meaning
- Revisiting Knowledge Transfer for Training Object Class Detectors (CVPR 2018 · similarity 0.90)
- Mixed Supervised Object Detection by Transferring Mask Prior and Semantic Similarity (NeurIPS 2021 · similarity 0.90)
- Instance-Aware, Context-Focused, and Memory-Efficient Weakly Supervised Object Detection (CVPR 2020 · similarity 0.89)
- Weakly Supervised Knowledge Transfer with Probabilistic Logical Reasoning for Object Detection (ICLR 2023 · similarity 0.89)
- W2F: A Weakly-Supervised to Fully-Supervised Framework for Object Detection (CVPR 2018 · similarity 0.88)
- Weakly Supervised Few-Shot Object Detection with DETR (AAAI 2024 · 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.