isl-org/unifi3d

[TMLR 2025] Unifi3D: A Study on 3D Representations for Generation and Reconstruction in a Common Framework

31
/ 100
Emerging

This project offers a unified framework for evaluating and comparing different 3D representations used in generative AI. It takes various 3D data formats like point clouds or SDFs and helps researchers understand which representation performs best for both generating new 3D models and accurately reconstructing existing ones. Scientists and engineers working on advanced 3D content creation would find this useful for benchmarking novel techniques.

Use this if you are a researcher or engineer who needs to benchmark and understand the trade-offs between different 3D data representations for generative tasks like creating new 3D models or reconstructing shapes.

Not ideal if you are looking for an off-the-shelf application to simply generate 3D models without needing to delve into the underlying representation performance.

3D-reconstruction generative-AI 3D-model-benchmarking computational-design computer-graphics-research
No Package No Dependents
Maintenance 6 / 25
Adoption 7 / 25
Maturity 15 / 25
Community 3 / 25

How are scores calculated?

Stars

41

Forks

1

Language

Python

License

Apache-2.0

Last pushed

Dec 17, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/generative-ai/isl-org/unifi3d"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.