AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Hierarchical Domain-Adapted Feature Learning for Video Saliency Prediction

Giovanni Bellitto, Federica Proietto Salanitri, Simone Palazzo, Francesco Rundo, Daniela Giordano, Concetto Spampinato

IJCV 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Geometric alignment and synthetic-to-real adaptation · 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 Conspicuity-net 3 to the encoder-decoder baseline (ablation)HD2S · DHF1K valNSS: 2.602 -> 2.612Improved
add Conspicuity-net 2 (ablation)HD2S · DHF1K valNSS: 2.612 -> 2.699Improved
add Conspicuity-net 1 (ablation)HD2S · DHF1K valNSS: 2.699 -> 2.743Improved
add multi-level loss on conspicuity maps (ablation)HD2S · DHF1K valNSS: 2.743 -> 2.806Improved
use only the first conspicuity net instead of the third (ablation)HD2S single-branch · DHF1K valNSS: 2.191 vs 2.663Mixed
add hierarchical gradient reversal domain adaptation (source DHF1K)HD2S · DHF1K to UCF SportsNSS: 2.483 -> 2.514Improved
add hierarchical gradient reversal domain adaptation (source UCF Sports)HD2S · UCF Sports to DHF1KNSS: 2.237 -> 2.160Hurt
add hierarchical gradient reversal domain adaptation (source Hollywood2)HD2S · Hollywood2 to UCF SportsNSS: 2.476 -> 2.389Hurt
add domain adaptation on DHF1K test benchmarkHD2S · DHF1K testNSS: 2.781 -> 2.709Hurt
add domain-specific learning modulesHD2S · DHF1K testNSS: 2.781 -> 2.812Improved
add domain-specific learning modulesHD2S · UCF SportsAUC-J: 0.913 -> 0.904Mixed
train on multiple datasets (multi-source) vs single sourceHD2S · DHF1K valNSS: 2.806 -> 2.811No clear effect

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