AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

COCO-Periph: Bridging the Gap Between Human and Machine Perception in the Periphery

Anne Harrington, Vasha DuTell, Mark Hamilton, Ayush Tewari, Simon Stent, William T. Freeman, Ruth Rosenholtz

ICLR 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Visual relationship and scene graph detection

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Evaluates how DNNs compare to humans under peripheral-vision constraints, using a COCO-based peripheral benchmark (COCO-Periph) built by modifying images to simulate peripheral vision.

Problem

Whether deep neural networks match human performance when human peripheral vision limits are imposed, and how such limits affect the human-machine perception gap.

Approach

Simulates peripheral vision by modifying images to mimic human peripheral degradation and evaluates DNNs against human performance on the resulting COCO-derived benchmark, aiming to see whether peripheral constraints narrow the gap between humans and machines.

Experimental setup

Datasets: COCO

Benchmarks: COCO-Periph

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