AMAP-ML/FE2E

[CVPR 2026] Beyond Generation: Advancing Image Editing Priors for Depth and Normal Estimation

46
/ 100
Emerging

Leverages a DiT-based architecture adapted from advanced image editing models (Step1X-Edit) via LoRA fine-tuning to perform monocular depth and surface normal estimation with zero-shot generalization. Achieves dense geometry prediction by transferring editing priors to geometry tasks, supporting multiple benchmark datasets (NYU-v2, KITTI, ScanNet, ETH3D) through unified inference pipelines with multi-GPU evaluation support.

195 stars.

No Package No Dependents
Maintenance 13 / 25
Adoption 10 / 25
Maturity 15 / 25
Community 8 / 25

How are scores calculated?

Stars

195

Forks

7

Language

Python

License

MIT

Last pushed

Mar 17, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/AMAP-ML/FE2E"

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