AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

VLT: Vision-Language Transformer and Query Generation for Referring Segmentation

Henghui Ding, Chang Liu, Suchen Wang, Xudong Jiang

TPAMI 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Referring expression comprehension · 178 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 transformer module instead of seven conv layers (ablation)VLT · RefCOCO test BIoU: 60.42 -> 65.24Improved
use Query Generation Module instead of language features as queries (ablation)VLT · RefCOCO test BIoU: 60.26 -> 65.24Improved
use Query Generation Module instead of learnt fixed queries (ablation)VLT · RefCOCO test BIoU: 58.60 -> 65.24Improved
increase queries from 1 to 16 (ablation)VLT · RefCOCO test BIoU: 57.34 -> 65.24Improved
increase queries from 16 to 32 (ablation)VLT · RefCOCO test BIoU: 65.24 -> 65.12No clear effect
add Query Balance Module (ablation)VLT · RefCOCO test BIoU: 63.80 -> 65.24Improved
use spatial-dynamic fusion instead of tile (ablation)VLT · RefCOCO test BIoU: 64.40 -> 65.24Improved
add contrastive learning (no masked samples) (ablation)VLT · RefCOCO test BIoU: 63.43 -> 64.70Improved
use masked contrastive learning instead of plain contrastive learning (ablation)VLT · RefCOCO test BIoU: 64.70 -> 65.24Improved
add masked contrastive learning, tested on expressions with words erased (ablation)VLT · RefCOCO test B (masked expressions)IoU: 59.53 -> 64.20Improved
add masked contrastive learning, trained on RefCOCO tested on RefCOCO+ val (ablation)VLT · RefCOCO to RefCOCO+ val (cross-dataset)IoU: 49.16 -> 52.35Improved
select masked words by probability p_m instead of random (ablation)VLT · RefCOCO to RefCOCO+ val (cross-dataset)IoU: 50.08 -> 52.35Improved

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