AI Paper Atlas

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

Top-Down Visual Attention From Analysis by Synthesis

Baifeng Shi, Trevor Darrell, Xin Wang

CVPR 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Scene graph generation · 39 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
add prior-conditioned top-down feedback to ViT-BVision-language model with ViT-B visual backbone · VQAv2 test-stdaccuracy: 67.92 -> 68.78Improved
add prior-conditioned top-down feedback to ViT-BVision-language model with ViT-B visual backbone · Flickr30K zero-shotIR@1: 42.40 -> 45.28Improved
replace top-down feedback with PerceiverIO token reweighting (ablation)ViT-B vision-language model · VQAv2 test-devaccuracy: 67.89 -> 67.87No clear effect
remove prior-based token reweighting but retain feedback (ablation)ViT-B vision-language model · VQAv2 test-devaccuracy: 67.89 -> 67.99Improved
use masked value-token reweighting instead of additive top-down input (ablation)ViT-B vision-language model · VQAv2 test-devaccuracy: 67.89 -> 67.53Hurt
add AbS feedback to ViT-TiViT-Ti · ImageNet-1Kaccuracy: 72.5 -> 74.1Improved
add AbS feedback to ViT-TiViT-Ti · ImageNet-CmCE: 71.1 -> 66.7Improved
add AbS feedback to ViT-SViT-S · ImageNet-Aaccuracy: 19.2 -> 24.3Improved
add AbS feedback to ViT-BViT-B · ImageNet-SKaccuracy: 43.3 -> 42.9Hurt
add AbSViT-B as the segmentation backboneViT-B segmentation backbone · PASCAL VOCsegmentation score: 80.1 -> 81.3Improved
add top-down input to query, key, and value rather than value only (ablation)AbSViT · ImageNet-1Kaccuracy: 74.1 -> 73.3Hurt
add top-down input to query, key, and value rather than value only (ablation)AbSViT · ImageNet-CmCE: 66.7 -> 68.0Hurt

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