atlas

Research map / Video / Video-language understanding

Divide and Conquer: Question-Guided Spatio-Temporal Contextual Attention for Video Question Answering

AAAI 2020 · Video › Video-language understanding · approach: Video question answering · 106 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
QueST vs HMEVideo QA · TGIF-QA FrameQAaccuracy: 53.8 -> 59.7Helped
QueST vs HMEVideo QA · TGIF-QA CountMSE: 4.02 -> 4.19Hurt
QueST vs HMEVideo QA · MSRVTT-QAaccuracy: 33.0 -> 34.6Helped
QueST vs HMEVideo QA · MSVD-QAaccuracy: 33.7 -> 36.1Helped
add video-guided word attention (ablation)QueST · TGIF-QA FrameQAaccuracy: 58.5 -> 59.7Helped
add diversity loss (ablation)QueST · TGIF-QA Transitionaccuracy: 79.3 -> 81.0Helped
add diversity loss (ablation)QueST · TGIF-QA CountMSE: 4.17 -> 4.19No clear effect
replace ST-VQA attention with contextual attention blocks (ablation)QueST · TGIF-QA FrameQAaccuracy: 56.6 -> 59.0Helped
multi-scale temporal CAB instead of single-scale (ablation)QueST · TGIF-QA Actionaccuracy: 74.1 -> 75.9Helped

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