A professional photo aesthetics evaluation platform powered by AI visual analysis, offering deep analysis of your work across composition, lighting, color, technical quality, and more.
Supports high-resolution photos up to 80MB. Oversized images are automatically turned into a 6000px-long-edge diagnosis copy.
Upload a photo and let Luoshen analyze it professionally across composition, lighting, color, and technical quality.
A full aesthetic assessment across composition, lighting, color, technical quality, and distractions.
Get concrete improvement advice based on the diagnosis to raise the aesthetic quality of your work.
About Luoshen
Luoshen is built on the multimodal large language model aesthetics framework proposed by the MIPL Lab at the Wangxuan Institute of Computer Technology, Peking University. Combined with Alibaba Qwen multimodal models and strengthened through training on 200,000 photography works, it focuses on aesthetic understanding, guidance, and cropping for photographic images.
It does more than judge whether a photo looks good. It identifies concrete issues in composition, lighting, color, and subject expression, then gives actionable ways to improve.
It helps find more balanced and visually appealing crop options, with clear aesthetic reasons such as stronger subject emphasis, cleaner edges, and steadier visual balance.
This site follows the aesthetic diagnosis approach of the Wangxuan Institute and provides fast photo analysis. You can optionally generate an AI expert interpretation on the report page. Documentary works should remain truthful. Removing real elements, compositing scenes, or fabricating shooting information is not recommended.