AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Target-Absent Human Attention

Zhibo Yang, Sounak Mondal, Seoyoung Ahn, Gregory Zelinsky, Minh Hoai, Dimitris Samaras

ECCV 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Human perception modeling · 30 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 auxiliary detection loss (ablation)FFMs + IQ-Learn · COCO-Search18 target-absentSemSS: 0.476 -> 0.516Improved
add auxiliary detection loss (ablation)FFMs + IQ-Learn · COCO-Search18 target-absentSS(2): 0.545 -> 0.537Hurt
use FFMs instead of DCBs (ablation)IQ-Learn · COCO-Search18 target-absentSemSS: 0.508 -> 0.516Improved
use FFMs instead of CFI (ablation)IQ-Learn · COCO-Search18 target-absentSemSS: 0.504 -> 0.516Improved
use FFMs instead of FPN P2 features (ablation)IQ-Learn · COCO-Search18 target-absentcIG: 0.018 -> 0.729Improved
use foveated (multi-resolution) FFMs instead of binary masks (ablation)IQ-Learn · COCO-Search18 target-absentSemSS: 0.510 -> 0.516Improved
replace IRL state with FFMs and IQ-Learnscanpath predictor · COCO-Search18 target-absentSemSS: 0.476 -> 0.516Improved
train individual model instead of group modelFFMs + IQ-Learn · COCO-Search18 target-absent (full scanpath)sequence scoreImproved
train individual model instead of group modelFFMs + IQ-Learn · COCO-Search18 target-absent (first 4 fixations)sequence scoreHurt
train individual model instead of group modelFFMs + IQ-Learn · COCO-Search18 target-absentMAE of scanpath lengthImproved
apply method to target-present search vs Chen et al.FFMs + IQ-Learn · COCO-Search18 target-presentSemSS: 0.572 vs 0.562Hurt
apply method to target-present search vs Chen et al.FFMs + IQ-Learn · COCO-Search18 target-presentSS: 0.445 -> 0.451Improved

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