PhotoAgent: Agentic Photo Editing with Exploratory Visual Aesthetic Planning
This work tackles the problem of automating complex, multi-step image editing for general users, reducing the need for detailed manual instruction design.
This paper addresses the challenge of autonomous image editing by introducing PhotoAgent, a system that uses explicit aesthetic planning to perform multi-step editing actions. PhotoAgent formulates editing as a long-horizon decision-making problem, reasoning over user intent and refining results through closed-loop execution with memory and visual feedback. The system consistently improves instruction adherence and visual quality compared to baseline methods.
With the recent fast development of generative models, instruction-based image editing has shown great potential in generating high-quality images. However, the quality of editing highly depends on carefully designed instructions, placing the burden of task decomposition and sequencing entirely on the user. To achieve autonomous image editing, we present PhotoAgent, a system that advances image editing through explicit aesthetic planning. Specifically, PhotoAgent formulates autonomous image editing as a long-horizon decision-making problem. It reasons over user aesthetic intent, plans multi-step editing actions via tree search, and iteratively refines results through closed-loop execution with memory and visual feedback, without requiring step-by-step user prompts. To support reliable evaluation in real-world scenarios, we introduce UGC-Edit, an aesthetic evaluation benchmark consisting of 7,000 photos and a learned aesthetic reward model. We also construct a test set containing 1,017 photos to systematically assess autonomous photo editing performance. Extensive experiments demonstrate that PhotoAgent consistently improves both instruction adherence and visual quality compared with baseline methods. The project page is https://mdyao.github.io/PhotoAgent/.