Qing Qu

Discover a new method for image restoration using self-guided reconstruction.

A new method improves image restoration through innovative diffusion model frameworks.

New method enhances image restoration by reducing noise and preserving details.

This article explores overparameterization and its impact on model training efficiency.

LOCO Edit allows precise image edits without additional training.

ViDiDi enhances video learning through efficient use of unlabeled data.

Discover how models bridge the gap between different data types.

Learn how to prevent model collapse in generative models using real data.