baegwangbin/IronDepth

[BMVC 2022] IronDepth: Iterative Refinement of Single-View Depth using Surface Normal and its Uncertainty

36
/ 100
Emerging

This project helps improve the quality of 3D depth maps generated from a single 2D image, making them more precise and detailed. It takes a standard color image along with an initial, rough depth map and a surface normal prediction, and outputs a highly refined, accurate depth map. This is useful for professionals working in 3D reconstruction, computer vision, and robotics who need high-fidelity spatial data.

182 stars. No commits in the last 6 months.

Use this if you need to significantly enhance the accuracy and detail of depth information derived from single images, especially when building 3D models or enabling machines to perceive their environment better.

Not ideal if you already have access to high-quality 3D sensor data like LiDAR or stereo vision, as this tool focuses on refining single-view depth estimations.

3D-reconstruction robotics computer-vision spatial-mapping augmented-reality
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

182

Forks

10

Language

Python

License

MIT

Last pushed

Apr 29, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/baegwangbin/IronDepth"

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