Problem
Research questions and friction points this paper is trying to address.
image restoration
generative AI
unified model
prompt design
model generalization
Can Nano Banana 2 Replace Traditional Image Restoration Models? An Evaluation of Its Performance on Image Restoration Tasks
🤖 AI Summary
This work investigates whether general-purpose generative image editing models can serve as a unified alternative to traditional task-specific restoration methods. Building upon NanoBanana2, we systematically evaluate its restoration performance across diverse degradation types and complex real-world scenarios, introducing a concise prompting strategy that incorporates explicit fidelity constraints. Experimental results demonstrate that the proposed approach surpasses state-of-the-art specialized models in full-reference metrics while achieving competitive perceptual quality. Notably, it excels in challenging cases such as small faces, dense crowds, and severely degraded inputs, underscoring the strong generalization capability of modern generative models for image restoration tasks.
📝 Abstract
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.
Innovation
Methods, ideas, or system contributions that make the work stand out.
image restoration
generative AI
prompt engineering
unified model
perceptual quality
🔎 Similar Papers
No similar papers found.
