3D Vision & Generative AI

Digit3D: 3D Multimodal MNIST Benchmark

Transforming 2D MNIST into rich 3D geometries (watertight meshes, sparse SDFs, and 6D normal-oriented point clouds) with continuous flow matching generative pipelines.

Khoi DO
Conquer3D Research & Open-Source Geometry Engine
Code Conquer3D Engine Overview Unconditional Generation Class-Conditional Generation Image-Conditional Generation Arbitrary-Resolution Generation Sparse Voxel Reconstruction Image-Conditional Mesh Generation PointCloud-Conditional Mesh Generation Limitations
Experiment 1

Unconditional Point Cloud Generation

Generating continuous 3D digit geometries spontaneously from pure Gaussian noise $\mathbf{x}_0 \sim \mathcal{N}(0, \mathbf{I})$ using the Mean Flow velocity matching model.

Sample #000
Sample #001
Sample #002
Sample #003
Sample #004
Sample #005
Sample #006
Sample #007
Sample #008
Sample #009

Drag to rotate 3D point cloud • Scroll to zoom • Shaded by continuous surface normal vectors