88‐5: <i>Invited Paper:</i> Enabling the Under‐Display Camera: Solving Video Quality Using AI Within the ISP

Under‐display cameras (UDCs) introduce significant image quality challenges due to optical distortions, increased noise, and reduced light transmission. This paper presents an AIdriven image restoration method designed for real‐time execution within an Image Signal Processor (ISP). The approach leverages a recurrent neural network (RNN)‐based denoiser with a lightweight architecture optimized for embedded platforms, enabling efficient noise reduction and distortion correction. We evaluate our method using objective image quality metrics such as Signal‐to‐Noise Ratio (SNR), Peak Signal‐to‐Noise Ratio (PSNR), and Structural Similarity Index (SSIM) before and after processing. The AI‐enhanced pipeline significantly improves image clarity while operating within power and computational constraints. Comparative analysis with traditional denoising and deblurring algorithms demonstrates this method's advantages in preserving fine details while suppressing structured noise.