3 ms·
Depth Anything: Unleashing the power of large-scale unlabeled data
- GaggiX 3y agoPrevious thread: https://news.ycombinator.com/item?id=39091981 https://news.ycombinator.com/item?id=39091981
- liuliu 3y agoNote that this model works well at higher resolution too (in contrast with MiDaS family) given the depth head is convolution and the DINO v2 base can interpolate positional encoding to generate higher-resolution features: https://x.com/drawthingsapp/status/1757486115802726744 https://x.com/drawthingsapp/status/1757486115802726744