AI Paper Atlas

Research map / Video / Video-language understanding

TL;DW? Summarizing Instructional Videos with Task Relevance & Cross-Modal Saliency

Medhini Narasimhan, Arsha Nagrani, Chen Sun, Michael Rubinstein, Trevor Darrell, Anna Rohrbach, Cordelia Schmid

ECCV 2022 · Video › Video-language understanding · approach: Dense video captioning

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 cross-modal saliency to task relevance for pseudo summaries (ablation)IV-Sum · WikiHow Summaries valF-Score: 64.1 -> 67.9Improved
add task relevance to cross-modal saliency for pseudo summaries (ablation)IV-Sum · WikiHow Summaries valF-Score: 61.0 -> 67.9Improved
pseudo summaries instead of step-localization annotations (ablation)IV-Sum · WikiHow Summaries valF-Score: 57.6 -> 66.8Improved
add ASR transcript input (ablation)IV-Sum · WikiHow Summaries testF-Score: 65.2 -> 67.3Improved
fine-tune S3D mixed_5 layers instead of freezing (ablation)IV-Sum · WikiHow Summaries valF-Score: 65.8 -> 67.9Improved
24-layer transformer instead of 12 (ablation)IV-Sum · WikiHow Summaries valF-Score: 66.7 -> 67.9Improved
transformer instead of MLP scorer (ablation)IV-Sum · WikiHow Summaries valF-Score: 32.1 -> 67.9Improved
add diversity loss (ablation)IV-Sum · WikiHow Summaries valF-Score: 67.9 -> 61.2Hurt
add reconstruction loss (ablation)IV-Sum · WikiHow Summaries valF-Score: 67.9 -> 67.6No clear effect
MIL-NCE S3D segment features instead of CLIP frame features (ablation)IV-Sum · WikiHow Summaries testF-Score: 62.0 -> 67.3Improved
train CLIP-It on pseudo summaries instead of generic datasetsCLIP-It with ASR · WikiHow Summaries testF-Score: 27.2 -> 61.8Improved
IV-Sum trained on pseudo summaries vs pseudo summary generation itselfIV-Sum · WikiHow Summaries subset (15 tasks)F-Score: 38.0 -> 42.0Improved

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