3 ms·DA-Clip: Controlling Vision-Language Models for Universal Image Restoration1 points by mvoodarla 3y ago