AI Paper Atlas

Research map / Video / Video-language understanding

Sketch, Ground, and Refine: Top-Down Dense Video Captioning

Chaorui Deng, Shizhe Chen, Da Chen, Yuan He, Qi Wu

CVPR 2021 · Video › Video-language understanding · approach: Dense video captioning · 69 citations (OpenAlex)

Check 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
generate a video-level paragraph first, then ground and refine its event sentences (ablation)SGR · ActivityNet Captioning validationMETEOR: 7.79 -> 9.07Improved
generate a video-level paragraph first, then ground and refine its event sentences (ablation)SGR · YouCook2METEOR: 3.67 -> 4.35Improved
add caption refinement without a second grounding stage (ablation)SG · ActivityNet Captioning validation (C3D)METEOR: 7.79 -> 8.85Improved
add a second grounding stage using the refined captions (ablation)SG with refine caption · ActivityNet Captioning validation (C3D)METEOR: 8.85 -> 9.07Improved
add caption refinement without a second grounding stage (ablation)SG · ActivityNet Captioning validation (TSN)METEOR: 8.27 -> 9.13Improved
add a second grounding stage using the refined captions (ablation)SG with refine caption · ActivityNet Captioning validation (TSN)METEOR: 9.13 -> 9.37Improved
add caption refinement without a second grounding stage (ablation)SG · YouCook2METEOR: 3.67 -> 3.96Improved
add a second grounding stage using the refined captions (ablation)SG with refine caption · YouCook2METEOR: 3.96 -> 4.35Improved
use the sentence-level advantage in RL refinement rewards (ablation)SGR refinement training · ActivityNet CaptioningMETEORImproved
use the paragraph-level advantage in RL refinement rewards (ablation)SGR refinement training · ActivityNet CaptioningMETEORImproved
use both hierarchical advantage functions instead of cross-entropy refinement alone (ablation)SGR refinement training · ActivityNet CaptioningMETEORImproved
attend to the aligned video segment during caption refinement (ablation)SGR dual-path cross attention · ActivityNet CaptioningMETEORImproved

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