Lotus: A diffusion-based visual foundation model for dense geometry prediction.
Predicting dense geometry in computer vision involves estimating properties such as depth and surface normals for each pixel in an ...
Predicting dense geometry in computer vision involves estimating properties such as depth and surface normals for each pixel in an ...
Autoregressive imaging models have traditionally relied on vector-quantized representations, which introduces several important challenges. The vector quantization process requires a ...
Despite the amazing advances and achievements in the field of technology, classical diffusion models still face challenges in imaging, particularly ...
You may have missed a big breakthrough in the ML weather forecasting revolution over the holidays: GenCast – Google DeepMind's ...
Physics-based character animation, a field at the intersection of computer graphics and physics, aims to create realistic and responsive character ...
How can the effectiveness of vision transformers be harnessed in diffusion-based generative learning? This NVIDIA paper presents a novel model ...
In the field of text-to-music synthesis, the quality of the generated content has been advancing, but the controllability of the ...