AI Paper Atlas

Research map / Video / Video-language understanding

Weakly-supervised Video Summarization using Variational Encoder-Decoder and Web Prior

Sijia Cai, Wangmeng Zuo, Larry S. Davis, Lei Zhang

ECCV 2018 · Video › Video-language understanding · approach: Dense video captioning · 89 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 30 web videos per topic under weak supervision (ablation)VESD · CoSumtop-5 mAP: 0.616 -> 0.684Improved
add 60 web videos per topic under weak supervision (ablation)VESD · CoSumtop-5 mAP: 0.616 -> 0.701Improved
add 30 web videos per topic under weak supervision (ablation)VESD · TVSumtop-5 mAP: 0.352 -> 0.407Improved
add 60 web videos per topic under weak supervision (ablation)VESD · TVSumtop-5 mAP: 0.352 -> 0.423Improved
add 30 web videos per topic under strong supervision (ablation)VESD · CoSumtop-5 mAP: 0.712 -> 0.755Improved
increase web videos from 30 to 60 per topic under strong supervision (ablation)VESD · CoSumtop-5 mAP: 0.755 -> 0.764Improved
add 30 web videos per topic under strong supervision (ablation)VESD · TVSumtop-5 mAP: 0.437 -> 0.481Improved
increase web videos from 30 to 60 per topic under strong supervision (ablation)VESD · TVSumtop-5 mAP: 0.481 -> 0.498Improved
add importance-score supervision to topic labels (ablation)VESD · CoSumtop-5 mAP: 0.616 -> 0.712Improved
add importance-score supervision to topic labels (ablation)VESD · TVSumtop-5 mAP: 0.352 -> 0.437Improved
replace GoogLeNet features with ResNet101 (ablation)VESD · CoSumtop-5 mAP: 0.755 -> 0.767Improved
replace GoogLeNet features with C3D (ablation)VESD · CoSumtop-5 mAP: 0.755 -> 0.765Improved

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