AI Paper Atlas

Research map / Video / Video-language understanding

Location-Aware Graph Convolutional Networks for Video Question Answering

Deng Huang, Peihao Chen, Runhao Zeng, Qing Du, Mingkui Tan, Chuang Gan

AAAI 2020 · Video › Video-language understanding · approach: Video question answering · 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 a location-aware object graph with GCN reasoningVideo QA · TGIF-QAaccuracy: 70.58 -> 74.32Improved
add detected object features (ablation)global-frame baseline · TGIF-QAaccuracy: 70.58 -> 72.82Improved
add GCN object-relation reasoning (ablation)object-feature variant · TGIF-QAaccuracy: 72.82 -> 74.10Improved
replace GCNs with two FC layers (ablation)object-feature variant · TGIF-QAaccuracy: 74.10 -> 72.96Hurt
replace GCNs with a 2-layer LSTM (ablation)object-feature variant · TGIF-QAaccuracy: 74.10 -> 72.65Hurt
add spatial and temporal location features (ablation)object-feature GCN variant · TGIF-QAaccuracy: 74.10 -> 74.32Improved
use temporal location only instead of both location types (ablation)L-GCN · TGIF-QAaccuracy: 74.32 -> 73.75Hurt
use spatial location only instead of both location types (ablation)L-GCN · TGIF-QAaccuracy: 74.32 -> 73.58Hurt
increase GCN depth from 2 to 3 layers (ablation)L-GCN · TGIF-QA Actionaccuracy: 74.32 -> 74.32No clear effect
increase GCN depth from 2 to 4 layers (ablation)L-GCN · TGIF-QA Actionaccuracy: 74.32 -> 73.97Hurt
use 2 objects per frame instead of 5 (ablation)L-GCN · TGIF-QA Actionaccuracy: 74.32 -> 74.11Hurt
use 10 objects per frame instead of 5 (ablation)L-GCN · TGIF-QA Actionaccuracy: 74.32 -> 74.01Hurt

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