AI Paper Atlas

Research map / Video / Video-language understanding

Less is More: Picking Informative Frames for Video Captioning

Yangyu Chen, Shuhui Wang, Weigang Zhang, Qingming Huang

ECCV 2018 · Video › Video-language understanding · approach: Encoder-decoder video captioning · 226 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 language reward to visual-diversity-only PickNet (ablation)PickNet · MSVD testCIDEr: 75.6 -> 76.0Improved
add visual diversity reward to language-only PickNet (ablation)PickNet · MSVD testCIDEr: 74.7 -> 76.0Improved
add language reward to visual-diversity-only PickNet (ablation)PickNet · MSVD testBLEU@4: 43.6 -> 46.1Improved
add visual diversity reward to language-only PickNet (ablation)PickNet · MSVD testBLEU@4: 49.9 -> 46.1Hurt
replace all-30-frame encoding with learned frame picking (about 6 frames)Encoder-Decoder baseline · MSVD testCIDEr: 69.4 -> 76.0Improved
replace random frame selection with PickNet pickingEncoder-Decoder · MSVD testCIDEr: 49.2 -> 76.0Improved
replace k-means frame selection with PickNet pickingEncoder-Decoder · MSVD testCIDEr: 70.9 -> 76.0Improved
replace all-30-frame encoding with PickNet picking (about 6 frames)Encoder-Decoder baseline · MSVD testrelative running time: 5x -> 1xImproved
replace all-30-frame encoding with PickNet picking (about 8 frames)Encoder-Decoder baseline · MSR-VTT testCIDEr: 41.2 -> 42.1Improved
add language reward to visual-diversity-only PickNet (ablation)PickNet · MSR-VTT testCIDEr: 41.0 -> 42.1Improved
add visual diversity reward to language-only PickNet (ablation)PickNet · MSR-VTT testCIDEr: 41.9 -> 42.1Improved
replace random frame selection with PickNet pickingEncoder-Decoder · MSR-VTT testCIDEr: 32.6 -> 42.1Improved

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