AI Paper Atlas

Research map / Video / Video-language understanding

Lightweight Attentional Feature Fusion: A New Baseline for Text-to-Video Retrieval

Fan Hu, Aozhu Chen, Ziyue Wang, Fangming Zhou, Jianfeng Dong, Xirong Li

ECCV 2022 · Video › Video-language understanding · approach: Text-video retrieval · 51 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
replace MHSA with LAFF block (ablation)text-video retrieval, 5 text features · MSR-VTT MV-test3kmAP: 0.294 -> 0.358Improved
add the attentional layer to Attention-free fusion (LAFF) (ablation)feature fusion · MSR-VTT MV-test3kmAP: 0.326 -> 0.358Improved
replace concatenation with Attention-free fusion (ablation)feature fusion · MSR-VTT MV-test3kmAP: 0.247 -> 0.326Improved
add multi-loss to MHSA (ablation)MHSA fusion · MSR-VTT MV-test3kmAP: 0.294 -> 0.306Improved
add BERT text feature to concatenation (W2VV++) (ablation)concatenation fusion · MSR-VTT MV-test3kmAP: 0.310 -> 0.247Hurt
add BERT text feature to LAFF (ablation)LAFF · MSR-VTT MV-test3kmAP: 0.358 -> 0.358No clear effect
add CLIP text feature to LAFF (ablation)LAFF · MSR-VTT MV-test3kmAP: 0.276 -> 0.358Improved
use combined per-space loss instead of single loss (ablation)LAFF · MSR-VTT MV-test3kmAP: 0.324 -> 0.358Improved
increase number of common spaces h from 1 to 8 (ablation)LAFF · MSR-VTT MV-test3kmAP: 0.352 -> 0.358Improved
increase h from 8 to 16 (ablation)LAFF · MSR-VTT MV-test3kmAP: 0.358 -> 0.356No clear effect
keep only top-3 video and text features chosen by LAFF weights (ablation)LAFF · MSR-VTT MV-test3kmAP: 0.358 -> 0.353No clear effect
use multi-level LAFF-ml instead of LAFFLAFF · MSR-VTT MV-test3kR@1: 28.0 -> 29.1Improved

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