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 2

Class-Conditional 3D Generation (Digits 0–9)

Steering 3D shape synthesis towards explicit digit classes. Point Transformer conditioning vectors incorporate learnable class embeddings with Classifier-Free Guidance (CFG, scale = 2.0).

0 Class 0
1 Class 1
2 Class 2
3 Class 3
4 Class 4
5 Class 5
6 Class 6
7 Class 7
8 Class 8
9 Class 9

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