NVIDIA Kaolin library provides a PyTorch API for working with a variety of 3D representations and includes a growing collection of GPU-optimized operations such as modular differentiable rendering, ...
FasterViT achieves a new SOTA Pareto-front in terms of Top-1 accuracy and throughput without extra training data ! We introduce a new self-attention mechanism, denoted as Hierarchical Attention (HAT), ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback