Can Nano Banana 2 Replace Traditional Image Restoration Models? An Evaluation of Its Performance on Image Restoration Tasks
This work addresses the problem of developing unified image restoration solutions for researchers and practitioners, but it is incremental as it focuses on evaluating an existing model rather than proposing a new method.
The paper evaluated whether the general-purpose generative model Nano Banana 2 can replace traditional image restoration models, finding that with optimized prompts, it achieves superior full-reference metrics and competitive perceptual quality across diverse scenes and degradation types.
Recent advances in generative AI raise the question of whether general-purpose image editing models can serve as unified solutions for image restoration. In this work, we conduct a systematic evaluation of Nano Banana 2 for image restoration across diverse scenes and degradation types. Our results show that prompt design plays a critical role, where concise prompts with explicit fidelity constraints achieve the best trade-off between reconstruction accuracy and perceptual quality. Compared with state-of-the-art restoration models, Nano Banana 2 achieves superior performance in full-reference metrics while remaining competitive in perceptual quality, which is further supported by user studies. We also observe strong generalization in challenging scenarios, such as small faces, dense crowds, and severe degradations. However, the model remains sensitive to prompt formulation and may require iterative refinement for optimal results. Overall, our findings suggest that general-purpose generative models hold strong potential as unified image restoration solvers, while highlighting the importance of controllability and robustness. All test results are available on https://github.com/yxyuanxiao/NanoBanana2TestOnIR.