AI Paper Atlas

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

Contrastive Grouping With Transformer for Referring Image Segmentation

Jiajin Tang, Ge Zheng, Cheng Shi, Sibei Yang

CVPR 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 69 citations (OpenAlex)

This paper tried a common change: Add a contrastive loss

Add grouping with contrastive loss to assign tokens to distinct regions · RefCOCO val · oIoU: 66.46 -> 70.81 Improved

What 1,317 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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 CGFormer rather than LAVTReferring image segmentation · RefCOCO valoIoU: 72.73 -> 74.75Improved
Use CGFormer rather than LAVTReferring image segmentation · RefCOCO+ valoIoU: 62.14 -> 64.54Improved
Use CGFormer rather than LAVTReferring image segmentation · G-Ref val-UoIoU: 61.24 -> 64.68Improved
Use CGFormer rather than CEFNetReferring image segmentation · ReferIt testoIoU: 66.70 -> 73.36Improved
Add one linguistic-conditioned query token (ablation)RefCOCO segmentation baseline · RefCOCO valoIoU: 65.70 -> 66.39Improved
Use multiple query tokens instead of one (ablation)RefCOCO segmentation model · RefCOCO valoIoU: 66.39 -> 66.46No clear effect
Add grouping with contrastive loss to assign tokens to distinct regions (ablation)RefCOCO segmentation model with multiple tokens · RefCOCO valoIoU: 66.46 -> 70.81Improved
Add learnable Gumbel-Softmax hard assignment (ablation)RefCOCO model with grouping · RefCOCO valoIoU: 70.81 -> 72.44Improved
Add parallel multi-scale decoding (ablation)RefCOCO model with hard assignment · RefCOCO valoIoU: 72.44 -> 73.28Improved
Use Consecutive Decoder instead of parallel multi-scale querying (ablation)RefCOCO multi-scale model · RefCOCO valoIoU: 73.28 -> 74.75Improved
Use CGFormer rather than VLT with the same backbone and text encoderRefCOCO segmentation · RefCOCO valoIoU: 70.89 -> 74.75Improved
Replace cosine similarity with dot product (ablation)CGFormer grouping · RefCOCO valoIoU: 74.75 -> 73.37Hurt

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