AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Optimal visual search based on a model of target detectability in natural images

Shima Rashidi, Krista Ehinger, Andrew Turpin, Lars Kulik

NeurIPS 2020 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Simulation-based inference · 8 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 logistic regression with MLP classifier on Alexnet features (ablation)foveated detectability model · Detectability, 18 backgroundsMSE vs human detectability: 0.0978 -> 0.256Hurt
Replace Alexnet with VGG19 features (logistic regression) (ablation)foveated detectability model · Detectability, 18 backgroundsMSE vs human detectability: 0.0978 -> 0.257Hurt
Replace Alexnet with luminance histogram features (ablation)foveated detectability model · Detectability, 18 backgroundsMSE vs human detectability: 0.0978 -> 0.3404Hurt
Replace Alexnet with HMAX features (ablation)foveated detectability model · Detectability, 18 backgroundsMSE vs human detectability: 0.0978 -> 0.224Hurt
Replace Alexnet with Textons (oriented filter banks) (ablation)foveated detectability model · Detectability, 18 backgroundsMSE vs human detectability: 0.0978 -> 0.2767Hurt
Use model detectabilities instead of human (oracle) detectabilities in ideal observer (ablation)Bayesian ideal observer · Visual search, 18 backgroundsnumber of fixationsNo 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