ComplexBench-Edit: Benchmarking Complex Instruction-Driven Image Editing via Compositional Dependencies

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Chenglin, Zhou, Yucheng, Wang, Qianning, Wang, Zhe, Zhang, Kai
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913895291027456
author Wang, Chenglin
Zhou, Yucheng
Wang, Qianning
Wang, Zhe
Zhang, Kai
author_facet Wang, Chenglin
Zhou, Yucheng
Wang, Qianning
Wang, Zhe
Zhang, Kai
contents Text-driven image editing has achieved remarkable success in following single instructions. However, real-world scenarios often involve complex, multi-step instructions, particularly ``chain'' instructions where operations are interdependent. Current models struggle with these intricate directives, and existing benchmarks inadequately evaluate such capabilities. Specifically, they often overlook multi-instruction and chain-instruction complexities, and common consistency metrics are flawed. To address this, we introduce ComplexBench-Edit, a novel benchmark designed to systematically assess model performance on complex, multi-instruction, and chain-dependent image editing tasks. ComplexBench-Edit also features a new vision consistency evaluation method that accurately assesses non-modified regions by excluding edited areas. Furthermore, we propose a simple yet powerful Chain-of-Thought (CoT)-based approach that significantly enhances the ability of existing models to follow complex instructions. Our extensive experiments demonstrate ComplexBench-Edit's efficacy in differentiating model capabilities and highlight the superior performance of our CoT-based method in handling complex edits. The data and code are released at https://github.com/llllly26/ComplexBench-Edit.
format Preprint
id arxiv_https___arxiv_org_abs_2506_12830
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ComplexBench-Edit: Benchmarking Complex Instruction-Driven Image Editing via Compositional Dependencies
Wang, Chenglin
Zhou, Yucheng
Wang, Qianning
Wang, Zhe
Zhang, Kai
Computer Vision and Pattern Recognition
Text-driven image editing has achieved remarkable success in following single instructions. However, real-world scenarios often involve complex, multi-step instructions, particularly ``chain'' instructions where operations are interdependent. Current models struggle with these intricate directives, and existing benchmarks inadequately evaluate such capabilities. Specifically, they often overlook multi-instruction and chain-instruction complexities, and common consistency metrics are flawed. To address this, we introduce ComplexBench-Edit, a novel benchmark designed to systematically assess model performance on complex, multi-instruction, and chain-dependent image editing tasks. ComplexBench-Edit also features a new vision consistency evaluation method that accurately assesses non-modified regions by excluding edited areas. Furthermore, we propose a simple yet powerful Chain-of-Thought (CoT)-based approach that significantly enhances the ability of existing models to follow complex instructions. Our extensive experiments demonstrate ComplexBench-Edit's efficacy in differentiating model capabilities and highlight the superior performance of our CoT-based method in handling complex edits. The data and code are released at https://github.com/llllly26/ComplexBench-Edit.
title ComplexBench-Edit: Benchmarking Complex Instruction-Driven Image Editing via Compositional Dependencies
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.12830