Research map / Detection and segmentation / Open-set and few-shot detection
DETR-ViP: Detection Transformer with Robust Discriminative Visual Prompts
Bo Qian, Dahu Shi, Xing Wei
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 |
|---|---|---|---|
| add image-text contrastive alignment (ablation) | VIS-GDINO-T · COCO / LVIS-minival | AP: 21.1 -> 29.2; 17.2 -> 23.4 | Improved |
| add global prompt integration (ablation) | VIS-GDINO-T + alignment · COCO / LVIS-minival | AP: 29.2 -> 35.6; 23.4 -> 33.0 | Improved |
| add visual-textual prompt relation distillation (ablation) | DETR-ViP-T intermediate · COCO / LVIS-minival | AP: 35.6 -> 41.5; 33.0 -> 39.5 | Improved |
| add naive encoder fusion from Grounding DINO (ablation) | DETR-ViP-T intermediate · COCO / LVIS-minival | AP: 41.5 -> 41.3; 39.5 -> 39.1 | Hurt |
| use encoder selective fusion instead of naive encoder fusion (ablation) | DETR-ViP-T intermediate · COCO / LVIS-minival | AP: 41.3 -> 42.2; 39.1 -> 40.6 | Improved |
| add naive decoder fusion (ablation) | DETR-ViP-T intermediate · COCO / LVIS-minival | AP: 42.2 -> 40.8; 40.6 -> 25.5 | Hurt |
| add decoder selective fusion (ablation) | DETR-ViP-T intermediate · COCO / LVIS-minival | AP: 42.2 -> 43.2; 40.6 -> 41.1 | Improved |
| replace supervised contrastive loss with relation distillation loss (ablation) | DETR-ViP-T · COCO / LVIS-minival | AP: 42.6 -> 43.2; 40.4 -> 41.1 | Improved |
| raise distillation weight lambda from 1.0 to 10.0 (ablation) | DETR-ViP-T · LVIS | AP: 40.5 -> 41.1 | Improved |
| raise distillation weight lambda from 10.0 to 20.0 (ablation) | DETR-ViP-T · LVIS | AP: 41.1 -> 39.8 | Hurt |
| vary selective fusion threshold theta from 0.1 to 0.5 (ablation) | DETR-ViP-T · LVIS | AP: 41.1 -> 39.2 | Hurt |
| lower selective fusion threshold theta from 0.1 to 0.05 (ablation) | DETR-ViP-T · LVIS | AP: 41.1 -> 38.7 | Hurt |
Closest papers by meaning
- CP-DETR: Concept Prompt Guide DETR Toward Stronger Universal Object Detection (AAAI 2025 · similarity 0.90)
- Just a Few Glances: Open-Set Visual Perception with Image Prompt Paradigm (AAAI 2025 · similarity 0.89)
- Visual Diversity and Region-aware Prompt Learning for Zero-shot HOI Detection (NeurIPS 2025 · similarity 0.88)
- DiPEx: Dispersing Prompt Expansion for Class-Agnostic Object Detection (NeurIPS 2024 · similarity 0.88)
- Multi-Modal Prompting for Open-Vocabulary Video Visual Relationship Detection (AAAI 2024 · similarity 0.87)
- DoGA: Enhancing Grounded Object Detection via Grouped Pre-Training with Attributes (AAAI 2025 · 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.
