atlas

Research map / Video / Video-language understanding

Spatio-Temporal Graph for Video Captioning With Knowledge Distillation

CVPR 2020 · Video › Video-language understanding · approach: Encoder-decoder video captioning · 265 citations (OpenAlex)

Check 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
add object-aware logit distillation instead of using the scene branch alone (ablation)Two-branch video captioner · MSVDCIDEr: 86.0 -> 93.0Helped
fuse branches by feature concatenation instead of logit distillation (ablation)Two-branch video captioner · MSVDCIDEr: 79.3 -> 93.0Hurt
use L2 feature-distance distillation instead of logit distillation (ablation)Two-branch video captioner · MSVDCIDEr: 80.3 -> 93.0Hurt
add temporal graph connections to a spatial graph (ablation)Spatio-temporal graph captioner · MSVDCIDEr: 91.8 -> 93.0Helped
add spatial graph connections to a temporal graph (ablation)Spatio-temporal graph captioner · MSVDCIDEr: 92.1 -> 93.0Helped
use spatial and temporal prior-informed connections instead of uniform dense connections (ablation)Spatio-temporal graph captioner · MSVDCIDEr: 91.3 -> 93.0Helped

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