AI Paper Atlas

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

Mining Interpretable AOG Representations From Convolutional Networks via Active Question Answering

Quanshi Zhang, Jie Ren, Ge Huang, Ruiming Cao, Ying Nian Wu, Song Chun Zhu

TPAMI 2020 · Vision-language and multimodal › Visual question answering · approach: Attention and prior-robust VQA · 13 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 active question-answering to AOG learning (vs AOG without QA) (ablation)AOG · CUB200-2011 (head part)normalized distance: 0.1084 -> 0.0434Improved
add active QA with only 10 part annotations vs AOG w/o QA with 20 (ablation)AOG · CUB200-2011 (head part)normalized distance: 0.1084 -> 0.0626Improved
increase part annotations from 10 to 20 (ablation)Ours · CUB200-2011 (head part)normalized distance: 0.0626 -> 0.0434Improved
increase part annotations from 10 to 30 (ablation)Ours · ILSVRC 2013 DET Animal-Partnormalized distance: 0.1126 -> 0.0909Improved
increase part annotations from 20 to 30 (ablation)Ours · ILSVRC 2013 DET Animal-Partnormalized distance: 0.0965 -> 0.0909Improved
increase part annotations from 10 to 30 for Ours+fastRCNN (ablation)Ours+fastRCNN · PASCAL VOC Part (PCP)PCP (%): 33.0 -> 50.5Improved
add second-stage fine-tuning with object boxes to Fast-RCNN (ablation)Fast-RCNN · CUB200-2011 (head part)normalized distance: 0.3105 -> 0.1989Improved
add object-box fine-tuning to CNN-PDD (ablation)CNN-PDD · CUB200-2011 (head part)normalized distance: 0.2446 -> 0.2694Hurt
use RBF-SVM instead of linear SVM on fc7 features (ablation)fc7 features · CUB200-2011 (head part)normalized distance: 0.2786 -> 0.3360Hurt
add fast-RCNN refinement of part localization with 10 annotationsOurs · PASCAL VOC Part (PCP)PCP (%): 33.0 vs 34.5No clear effect
more part annotations grow AOG semantic-part children but not part templates (ablation)AOG · Pascal VOC Partaverage number of AOG node children (layer 1): 3.15 -> 15.83Mixed
use Ours (10 annotations) instead of Fast-RCNN (2 fts) with 10 annotationsFast-RCNN (2 fts) · Pascal VOC Part (neck)normalized distance: 0.309 -> 0.152Improved

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