nianticlabs/implicit-depth

[CVPR 2023] Virtual Occlusions Through Implicit Depth

35
/ 100
Emerging

This project helps create realistic 3D scene compositions by precisely estimating depth from images, even when objects overlap. It takes an input image sequence and, optionally, basic depth maps (like those from AR models) to produce detailed depth information. This is useful for researchers and developers working on virtual objects in real-world scenes.

No commits in the last 6 months.

Use this if you need highly accurate, per-pixel depth information for complex scenes to correctly handle virtual occlusions, especially in augmented reality or computer vision applications.

Not ideal if you're looking for a simple, out-of-the-box tool for general-purpose depth estimation without access to existing 3D scene data or developer expertise.

augmented-reality 3d-reconstruction computer-vision-research scene-understanding virtual-object-integration
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

87

Forks

5

Language

Python

License

Last pushed

May 09, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/nianticlabs/implicit-depth"

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