Research map / Detection and segmentation / Open-set and few-shot detection
Improved Visual-Semantic Alignment for Zero-Shot Object Detection
Shafin Rahman, Salman Khan, Nick Barnes
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 focal loss with polarity loss (ablation) | Our-FL-word · MS-COCO 65/15 ZSD | mAP: 10.80 -> 12.02 | Improved |
| replace focal loss with polarity loss (ablation) | Our-FL-word · MS-COCO 65/15 GZSD seen | mAP: 37.56 -> 33.28 | Hurt |
| add vocabulary metric learning (ablation) | Our-PL-word · MS-COCO 65/15 ZSD | mAP: 12.02 -> 12.62 | Improved |
| use word vectors during training (ablation) | Baseline RetinaNet · MS-COCO 65/15 ZSD | mAP: 8.48 -> 10.80 | Improved |
| use word vectors during training (ablation) | Baseline RetinaNet · MS-COCO 65/15 GZSD seen | mAP: 36.96 -> 37.56 | Hurt |
| add word vectors in box-regression subnet (ablation) | Ours · MS-COCO 65/15 ZSD | mAP | Improved |
| full method vs baseline | RetinaNet · MS-COCO 65/15 GZSD | harmonic mAP: 14.03 -> 18.18 | Improved |
| use small top-T seen semantics (T=5) instead of all (ablation) | Ours · MS-COCO | mAP | Improved |
| use polarity loss with balanced CE only (gamma=0) (ablation) | Ours · MS-COCO ZSD | mAP: around 10% mAP | No clear effect |
Closest papers by meaning
- Zero-Shot Object Detection (ECCV 2018 · similarity 0.92)
- Generalized Zero-Shot Recognition Based on Visually Semantic Embedding (CVPR 2019 · similarity 0.90)
- Class Semantic Attribute Perception Guided Zero-Shot Learning (AAAI 2025 · similarity 0.89)
- Visual and Semantic Prompt Collaboration for Generalized Zero-Shot Learning (CVPR 2025 · similarity 0.89)
- Progressive Semantic-Guided Vision Transformer for Zero-Shot Learning (CVPR 2024 · similarity 0.89)
- TPR: Topology-Preserving Reservoirs for Generalized Zero-Shot Learning (NeurIPS 2024 · similarity 0.88)
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.
