princeton-computational-imaging/HNDR
Official code repository for the paper: "The Implicit Values of A Good Hand Shake: Handheld Multi-Frame Neural Depth Refinement"
This project helps computer vision researchers refine depth perception in images captured with handheld cameras. By taking multiple slightly different images and their corresponding initial depth estimates and camera positions, it produces a more accurate, high-resolution 3D depth map. It's designed for researchers developing or evaluating advanced imaging techniques for applications like augmented reality or robotics.
No commits in the last 6 months.
Use this if you are a computer vision researcher needing to significantly improve the 3D depth accuracy of images captured with a moving handheld camera.
Not ideal if you need a real-time depth solution for live camera feeds or do not have access to initial depth estimates and camera pose data.
Stars
72
Forks
12
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Mar 27, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/princeton-computational-imaging/HNDR"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
cake-lab/HybridDepth
Official implementation for HybridDepth Model [WACV 2025, ISMAR 2024]
soubhiksanyal/RingNet
Learning to Regress 3D Face Shape and Expression from an Image without 3D Supervision
nianticlabs/monodepth2
[ICCV 2019] Monocular depth estimation from a single image
tinghuiz/SfMLearner
An unsupervised learning framework for depth and ego-motion estimation from monocular videos
ialhashim/DenseDepth
High Quality Monocular Depth Estimation via Transfer Learning