AI Paper Atlas

Research map / Detection and segmentation / CNN features

Personalized Saliency and Its Prediction

Yanyu Xu, Shenghua Gao, Junru Wu, Nianyi Li, Jingyi Yu A

TPAMI 2018 · Detection and segmentation › CNN features · approach: RGB salient object detection · 56 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
predict PSM as USM plus discrepancy (CNN-PIEF) instead of directly from RGBCNN-PIEF · PSM dataset, closed-setCC: 0.6109 -> 0.6532Improved
use discrepancy-based CNN-PIEF instead of per-observer MultiConvNets (ML-Net)ML-Net · PSM dataset, closed-setCC: 0.4310 -> 0.6368Improved
use shared Multi-task CNN instead of MultiConvNets (BMS)BMS · PSM dataset, closed-setCC: 0.5510 -> 0.6390Improved
use CNN-PIEF instead of MultiConvNets (SalNet)SalNet · PSM dataset, closed-setCC: 0.6397 -> 0.6771Improved
jointly predict USM and discrepancy (Baseline 1) instead of feeding predicted USMRGB-only baseline · PSM dataset, closed-setCC: 0.6109 -> 0.5703Hurt
replace predicted USM with ground-truth USMCNN-PIEF (SalNet) · PSM dataset, closed-setCC: 0.6771 -> 0.8651Improved
fine-tune SalNet on the 1000-image PSM training setSalNet · PSM dataset, closed-setCC: 0.6238 -> 0.5060Hurt
add supervision on middle layers (ablation)Multi-task CNN and CNN-PIEF · PSM dataset, closed-setCCImproved
insert person-specific filters at conv5 instead of conv7 (ablation)LDS based CNN-PIEF · PSM dataset, closed-set (insert position)CC: 0.6163 -> 0.6532Improved
insert person-specific filters at all of conv5-7 instead of conv5 only (ablation)LDS based CNN-PIEF · PSM dataset, closed-set (insert position)CC: 0.6532 -> 0.5867Hurt
train on 20 observers and transfer to unseen observers instead of training on the open-setSalNet based CNN-PIEF · PSM dataset, open-set (10 unseen observers)CC: 0.6863 -> 0.6468Hurt
use different methods for USM and discrepancy prediction (LDS USM, ML-Net discrepancy) (ablation)Multi-task CNN · PSM dataset, closed-setCC: 0.6463 -> 0.5351Hurt

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