AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

RefCLIP: A Universal Teacher for Weakly Supervised Referring Expression Comprehension

Lei Jin, Gen Luo, Yiyi Zhou, Xiaoshuai Sun, Guannan Jiang, Annan Shu, Rongrong Ji

CVPR 2023 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised semantic segmentation · 27 citations (OpenAlex)

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
Use anchor-text contrastive learning with filtered anchors instead of detected-region rankingRefCLIP · RefCOCO valIoU@0.5 accuracy: 38.08 -> 60.36Improved
Use RefCLIP to pseudo-label MAttNet trainingMAttNet · RefCOCO testBIoU@0.5 accuracy: 38.59 -> 71.27Improved
Remove scale selection and confidence filtering (ablation)RefCLIP · RefCOCO valIoU@0.5 accuracy: 33.71 -> 60.36Hurt
Select anchors from one scale (ablation)RefCLIP · RefCOCO valIoU@0.5 accuracy: 33.71 -> 50.75Improved
Add confidence-based anchor filtering (ablation)RefCLIP · RefCOCO valIoU@0.5 accuracy: 50.75 -> 53.30Improved
Add negative anchor augmentation (ablation)RefCLIP · RefCOCO valIoU@0.5 accuracy: 53.30 -> 60.36Improved
Use the 52 × 52 anchor scale (ablation)RefCLIP · RefCOCO valIoU@0.5 accuracy: 13 × 13: 60.36; 52 × 52: 0.00Hurt
Use the 13 × 13 anchor scale rather than 26 × 26 (ablation)RefCLIP · RefCOCO valIoU@0.5 accuracy: 11.23 -> 60.36Improved
Keep the top 20% of anchors by confidence (ablation)RefCLIP · RefCOCO valIoU@0.5 accuracy: 20.84 -> 59.31Improved
Keep only the top 5% of anchors by confidence (ablation)RefCLIP · RefCOCO valIoU@0.5 accuracy: 60.36 -> 48.46Hurt
Add data augmentation to pseudo-label student training (ablation)RealGIN · RefCOCO valIoU@0.5 accuracy: 57.36 -> 58.99Improved
Add EMA to student training with data augmentation (ablation)RealGIN · RefCOCO valIoU@0.5 accuracy: 58.99 -> 59.43Improved

Later papers that compared with this method

1 later accepted paper reports RefCLIP as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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