CompBench: Benchmarking Complex Instruction-guided Image Editing
This work addresses the need for more realistic and comprehensive evaluation in image editing for AI researchers, though it is incremental as it builds on existing benchmarking efforts.
The authors tackled the problem of oversimplified benchmarks for instruction-guided image editing by introducing CompBench, a large-scale benchmark designed for complex scenarios with fine-grained instructions, which exposed fundamental limitations in current models.
While real-world applications increasingly demand intricate scene manipulation, existing instruction-guided image editing benchmarks often oversimplify task complexity and lack comprehensive, fine-grained instructions. To bridge this gap, we introduce, a large-scale benchmark specifically designed for complex instruction-guided image editing. CompBench features challenging editing scenarios that incorporate fine-grained instruction following, spatial and contextual reasoning, thereby enabling comprehensive evaluation of image editing models' precise manipulation capabilities. To construct CompBench, We propose an MLLM-human collaborative framework with tailored task pipelines. Furthermore, we propose an instruction decoupling strategy that disentangles editing intents into four key dimensions: location, appearance, dynamics, and objects, ensuring closer alignment between instructions and complex editing requirements. Extensive evaluations reveal that CompBench exposes fundamental limitations of current image editing models and provides critical insights for the development of next-generation instruction-guided image editing systems. The dataset, code, and models are available in https://comp-bench.github.io/.