AI Paper Atlas

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

High-Quality Entity Segmentation and Grounding

Lu Qi, Yiwen Chen, T B Zhang, Xiangtai Li, Xu Yang, Bo Du, Ming-Hsuan Yang

TPAMI 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary semantic segmentation · 16 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
pretrain Mask2Former R-50 on COCO instead of ImageNetMask2Former R-50 · EntitySeg entity segmentationAP^e: 26.0 -> 35.2Improved
use a stronger backbone (Swin-T to Swin-L) with COCO pretrainingMask2Former · EntitySeg entity segmentationAP^e: 42.9 -> 46.2Improved
increase single-scale input resolution from (800,1333) to (2700,4500)Mask2Former · EntitySeg entity segmentationAP^e: 39.5 -> 40.9 (train time 2.4 -> 60.8 GPU days)Improved
add CropFormer full-image plus crop fusion with association module and batch decoder (ablation)Mask2Former · EntitySeg entity segmentationAP^e: 39.5 -> 41.0Improved
replace CropFormer fusion with video-level Mask2Former or VITA (ablation)crop fusion · EntitySeg entity segmentationAP^e: 39.7 and 39.9 vs 41.0 for CropFormerNo clear effect
pretrain on EntitySeg instead of COCO-Panoptic then fine-tune on ADE20KMask2Former · ADE20K transferENT AP^e: +1.5Improved
pretrain on EntitySeg instead of COCO-Panoptic then fine-tune on UVO videoMask2Former · UVO transferENT AP^e: 29.5 (+2.2)Improved
train on EntitySeg instead of COCO entity annotationsMask2Former Swin-L · In-the-wild avg AR@100AR@100: 52.23 -> 67.41Improved
scale GELLA language model from 7B to 13BGELLA · RefCOCO valcIoU: 76.1 -> 76.7No clear effect
scale GELLA language model from 7B to 13BGELLA · PNGAR: 69.8 -> 71.3Improved
scale GELLA language model from 7B to 13BGELLA · COCO panoptic segmentation (50 epochs)PQ: 56.4 -> 56.5No clear effect
train GELLA 50 epochs instead of 12GELLA-7B · COCO panoptic segmentationPQ: 53.1 -> 56.4Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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