AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

Weakly Supervised Relative Spatial Reasoning for Visual Question Answering

Pratyay Banerjee, Tejas Gokhale, Yezhou Yang, Chitta Baral

ICCV 2021 · Vision-language and multimodal › Visual question answering · approach: Vision-language model VQA evaluation · 20 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 weak supervision from 3D relative position estimation (15-bin classification) (ablation)LXMERT + SR · GQA-Valaccuracy: 59.40 -> 62.32Improved
add weak supervision from 2D object centroid estimation (regression) (ablation)LXMERT + SR · GQA-Valaccuracy: 59.40 -> 57.33Hurt
add weak supervision from 3D object centroid estimation (regression) (ablation)LXMERT + SR · GQA-Valaccuracy: 59.40 -> 58.28Hurt
add weak supervision from 2D object centroid estimation (15-bin classification) (ablation)LXMERT + SR · GQA-Valaccuracy: 59.40 -> 58.64Hurt
add weak supervision from 2D relative position regression (ablation)LXMERT + SR · GQA-Valaccuracy: 59.40 -> 59.85Improved
use early fusion of relative position vectors instead of late fusion (ablation)LXMERT + SR · 3D RPE 15-wayaccuracy: 69.45 -> 74.20Improved
add relative position vectors via early fusion (ablation)LXMERT + SR · 3D RPE 15-wayaccuracy: 52.80 -> 74.20Improved
add spatial pyramid image patches to early fusion (ablation)LXMERT + SR + Early Fusion · 3D RPE 15-wayaccuracy: 74.20 -> 81.47Improved
add spatial pyramid image patches to late fusion (ablation)LXMERT + SR + Late Fusion · 3D RPE 15-wayaccuracy: 69.45 -> 79.60Improved
add 9x9 grid patches to the patch set (ablation)Spatial pyramid patches · GQA-ValaccuracyNo clear effect
extract patch features with ResNet101 instead of ResNet-50 (ablation)Spatial pyramid patches · GQA-Valaccuracy: +0.05%Improved
add weakly supervised relative position estimation with early fusion and patchesLXMERT · GQA test-standardaccuracy: 60.34 -> 62.11Improved

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