Wei Li

PersonaDiff: diffusion-based style control for speech-synchronized 3D facial animation

Fft-guided multi-scale frequency enhancement for diffusion-based RAW-to-sRGB mapping

Experimental results on the MAI dataset demonstrate that DFEM-ISP attains a 0.79 dB PSNR enhancement, coupled with markedly improved high-frequency details and color consistency, thereby confirming the efficacy of FFT-driven frequency enhancement for RAW-to-sRGB reconstruction.

PDCFL: Prototype‐Driven Decoupled Clustered Federated Learning for Non‐IID Data

A prototype space optimization mechanism based on NT‐Xent that alleviates insufficient prototype discriminability by promoting intra‐cluster aggregation and increasing intercluster distance is devised and an adaptive model decoupling strategy is introduced to address the remaining intra‐cluster heterogeneity.