Fast and memory-efficient exact attention

An open survey on 3D Gaussian Splatting compression methods

UNet diffusion model in pure CUDA

Wave Function Collapse x Stable Diffusion, tile map generation with diffusion algorithm

Infinite Photorealistic Worlds using Procedural Generation

4M: Massively Multimodal Masked Modeling

On-device Inference of Diffusion Models for Apple Silicon

NDE: Neural Directional Encoding

LLM training in simple, raw C/CUDA

GeneFace++: Generalized and Stable Real-Time 3D Talking Face Generation; Official Code

A massively parallel, high-level programming language

Lumina-T2X is a unified framework for Text to Any Modality Generation

The most powerful and modular stable diffusion GUI, api and backend with a graph/nodes interface.

CUDA Core Compute Libraries

A little repo to render objaverse objects with blender

(CVPR 2024) Accelerating Neural Field Training via Soft Mining

GPU Ray Tracing Library

Tri-MipRF: Tri-Mip Representation for Efficient Anti-Aliasing Neural Radiance Fields, ICCV'23 (Oral, Best Paper Finalist)

Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)

A Metal re-implementation of tiny-cuda-nn's "learning a 2D image"

Source code for the CVPR'20 paper "Blindly Assess Image Quality in the Wild Guided by A Self-Adaptive Hyper Network"

An extremely fast Python linter and code formatter, written in Rust.

Differentiable Iso-Surface Extraction Package (DISO)

⚡ A Fast, Extensible Progress Bar for Python and CLI

Hackable and optimized Transformers building blocks, supporting a composable construction.

The repository for IEEE CVPR 2023 (A Light Weight Model for Active Speaker Detection)

ACM MM 2021: 'Is Someone Speaking? Exploring Long-term Temporal Features for Audio-visual Active Speaker Detection'

Out of time: automated lip sync in the wild