AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

ViTPrompt: Training-Free Prompt Refinement with Visual Tokens for Open-Vocabulary Detection

Yitong Qin, Lihua Zhou, Jiwei Wei, Ran Ran, Shiyuan He, Zeyu Ma, Shuaifeng Li, Nianxin Li, Heng Tao Shen

CVPR 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary object detection

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 changedWhereMeasuredEffect
add RoI-aligned visual tokens to the text prompt (max fusion) (ablation)ViTPrompt on Grounding DINO Swin-T · COCO-C (avg)mAP50: 25.57 -> 26.25Improved
add RoI-aligned visual tokens to the text prompt (max fusion) (ablation)ViTPrompt on Grounding DINO Swin-T · FoggyCityscapesmAP50: 28.89 -> 29.10Improved
two-stage visual token prompting for localizationGrounding DINO Swin-T · FoggyCityscapesmAP@0.5:0.95: 15.65 -> 19.02Improved
two-stage visual token prompting for localizationGrounding DINO Swin-T · COCO-C (avg)mAP@0.5:0.95: 15.27 -> 18.36Improved
increase top-K visual tokens beyond 10 (ablation)ViTPrompt · FoggyCityscapes / COCO-CmAP50: FoggyCityscapes 29.20 -> 28.76 across KNo clear effect
vary NMS IoU threshold between 0.2 and 0.8 (ablation)ViTPrompt · COCO-C and FoggyCityscapesmAP50: COCO-C 26.02 -> 26.30No clear effect
ViTPrompt versus BCA+ on PASCAL-CViTPrompt · PASCAL-C Swin-TmAP50: 40.67 -> 43.03Improved
ViTPrompt versus BCA+ on COCO-CViTPrompt · COCO-C Swin-BmAP50: 39.98 -> 42.28Improved
ViTPrompt versus BCA on Rider classViTPrompt · FoggyCityscapes Swin-T per-class RiderAP50: 5.61 -> 4.72Hurt
add second forward pass with visual tokensGrounding DINO Swin-T · Efficiency 600x600FLOPs (G): 145.64 -> 252.26Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy