nianticlabs/doubletake
[ECCV 2024] DoubleTake: Geometry Guided Depth Estimation
This project helps computer vision researchers and developers accurately estimate the depth of a scene from a series of ordinary color images, even when the camera moves. You provide multiple color images with information about their camera positions, and it produces a detailed depth map for each image. This is ideal for those working on 3D reconstruction, augmented reality, or robotics applications.
191 stars. No commits in the last 6 months.
Use this if you need to generate highly accurate depth maps from multiple posed RGB images, especially in scenarios where existing methods might struggle with geometric consistency.
Not ideal if you only have a single image and no camera pose information, or if you need to run this on very constrained hardware without dedicated GPUs.
Stars
191
Forks
12
Language
Python
License
—
Category
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/doubletake"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
vita-epfl/monoloco
A 3D vision library from 2D keypoints: monocular and stereo 3D detection for humans, social...
fangchangma/self-supervised-depth-completion
ICRA 2019 "Self-supervised Sparse-to-Dense: Self-supervised Depth Completion from LiDAR and...
nburrus/stereodemo
Small Python utility to compare and visualize the output of various stereo depth estimation algorithms
JiawangBian/sc_depth_pl
SC-Depth (V1, V2, and V3) for Unsupervised Monocular Depth Estimation ...
wvangansbeke/Sparse-Depth-Completion
Predict dense depth maps from sparse and noisy LiDAR frames guided by RGB images. (Ranked 1st...