AICuda★ 31
yassa9/dvlt.cu
DVLT on CUDA without dependencies
The project ports NVIDIA DVLT to pure CUDA/C++, allowing images to be fed in and a 3D point cloud together with camera poses to be obtained. Everything is packed into a single fast 5‑MB binary without Python.
// KEY FEATURES
- Generates a dense 3D point cloud from a sequence of images on GPU.
- Estimates camera poses (position and orientation) for each frame in real time.
- Works without external dependencies, a single ~5‑MB file, pure CUDA/C++.
- Requires no Python or additional libraries, everything in a single executable file.
#3d#3d-models#3d-reconstruction#cuda#cuda-cpp#cuda-kernels#cuda-programming#cutlass#inference#kernels#multi-view-stereo#transformer
Open on GitHub →New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram