Zero-1-to-3 on Hugging Face Accelerate

The implemention of the NeRF architecture from scratch that can be trained using synthetic data for 3D model rendering.

A TensorFlow implementation of Neural Radiance Fields (NeRF) for synthesizing novel views of complex 3D scenes from a sparse set of 2D images.

implementation of neural radiance fields for novel view synthesis and 3d reconstruction

Novel View Synthesis Using NeRF

Project page for "Self-Supervised Multimodal NeRF for Autonomous Driving"

A writing and worldbuilding workspace for writers and Dungeon Masters.

AI2-THOR to Blender Dataset Converter

IB3P

Single-image world synthesis using a generative panorama prior, orbit-conditioned spherical SfM, and Semantic 3D Gaussian scene reconstruction.

SNAP: Solving Neural Articulated Performances

Custom Implementation of NeRF: Neural Rendering Fields for View Synthesis

stable-virtual-camera 2D图片转3D视频 免安装一键启动整合包

A project where we train a Flow to go from pose distribution to image distribution and try to reverse it to estimate poses

Quantized 2D Gaussian Splatting for memory efficient 3D Reconstruction and Novel View Synthesis

From-scratch PyTorch 3DGS: cameras, covariance, SH, differentiable rendering, densification/pruning.

✭ MAGNETRON ™ ✭: Implementation of Google's Dreambooth (with Stable Diffusion model) instead of IMAGEN (for making MAGNETRON ™ TECHNOLOGY IMAGINATION PROXIA). In particular this can be used to mimic the appearance of subjects in a given reference set and synthesize novel renditions of them in different contexts.

Vision Based Algorithm For Generation of Novel Views

Real-Time View Synthesis with Multiplane Image Network using Multimodal Supervision