Salvato in:
Dettagli Bibliografici
Autori principali: Ghazanfari, Sara, Lin, Wei-An, Tian, Haitong, Yumer, Ersin
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2508.18159
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914058646585344
author Ghazanfari, Sara
Lin, Wei-An
Tian, Haitong
Yumer, Ersin
author_facet Ghazanfari, Sara
Lin, Wei-An
Tian, Haitong
Yumer, Ersin
contents Visually-guided image editing, where edits are conditioned on both visual cues and textual prompts, has emerged as a powerful paradigm for fine-grained, controllable content generation. Although recent generative models have shown remarkable capabilities, existing evaluations remain simple and insufficiently representative of real-world editing challenges. We present SpotEdit, a comprehensive benchmark designed to systematically assess visually-guided image editing methods across diverse diffusion, autoregressive, and hybrid generative models, uncovering substantial performance disparities. To address a critical yet underexplored challenge, our benchmark includes a dedicated component on hallucination, highlighting how leading models, such as GPT-4o, often hallucinate the existence of a visual cue and erroneously perform the editing task. Our code and benchmark are publicly released at https://github.com/SaraGhazanfari/SpotEdit.
format Preprint
id arxiv_https___arxiv_org_abs_2508_18159
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SpotEdit: Evaluating Visually-Guided Image Editing Methods
Ghazanfari, Sara
Lin, Wei-An
Tian, Haitong
Yumer, Ersin
Computer Vision and Pattern Recognition
Machine Learning
Visually-guided image editing, where edits are conditioned on both visual cues and textual prompts, has emerged as a powerful paradigm for fine-grained, controllable content generation. Although recent generative models have shown remarkable capabilities, existing evaluations remain simple and insufficiently representative of real-world editing challenges. We present SpotEdit, a comprehensive benchmark designed to systematically assess visually-guided image editing methods across diverse diffusion, autoregressive, and hybrid generative models, uncovering substantial performance disparities. To address a critical yet underexplored challenge, our benchmark includes a dedicated component on hallucination, highlighting how leading models, such as GPT-4o, often hallucinate the existence of a visual cue and erroneously perform the editing task. Our code and benchmark are publicly released at https://github.com/SaraGhazanfari/SpotEdit.
title SpotEdit: Evaluating Visually-Guided Image Editing Methods
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2508.18159