PulseAugur
EN
LIVE 18:04:29

AI Image Editors Fail to Accurately Simulate Optical Blur, Study Finds

By PulseAugur Editorial · [1 sources] ·

A pilot audit of image editing tools Gemini 3.1 Flash Image and GPT-image-2.5 reveals that they do not accurately simulate depth-dependent blur and aperture response as described by thin-lens optics. When asked to produce an f/1.4 effect, the blur added by these editors did not scale correctly with aperture changes or depth, unlike the results from Blender Cycles, a rendering engine used as a ground truth. Qualitative wording in prompts significantly impacted blur strength more than specific f-number requests, indicating a disconnect between user input and optical simulation. AI

IMPACT Highlights limitations in current AI image generation models regarding realistic optical simulation, suggesting areas for improvement in depth-dependent blur and aperture response.

RANK_REASON Research paper analyzing AI model capabilities. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

AI Image Editors Fail to Accurately Simulate Optical Blur, Study Finds

How we ranked this

Signal score
4 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Research paper analyzing AI model capabilities. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, product, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Same-day
Cluster formed today. Ranking reflects the current source set at time of score.

Full methodology in our editorial standards.

COVERAGE [1]

  1. arXiv cs.CV TIER_1 English(EN) · Zhihan Chen, Yuhuan Zhao, Yijie Zhu, Xinyu Yao, Mengcong Ren, Yuchen Sun, Yunqing Chen ·

    Do Image Editors Follow Depth-Dependent Blur and Aperture Response? A Rendered-Ground-Truth Pilot Audit

    arXiv:2610.09344v1 Announce Type: new Abstract: General image editors are asked to make a photo look as if it were taken at f/1.4, yet it is rarely checked whether the blur they add follows thin-lens optics. A physical aperture edit spreads blur across depth in thin-lens proporti…

RELATED TOPICS