2 ms·
Right there in the abstract: "Our approach works by predicting a dense motion flow field and a monocular depth map directly from a single motion-blurred image"
by skywal_l 2y ago
Right there in the abstract:
"Our approach works by predicting a dense motion flow field and a monocular depth map directly from a single motion-blurred image"