RuojinCai/doppelgangers
Doppelgangers: Learning to Disambiguate Images of Similar Structures
When working with many images of a real-world object or scene, it's common to use Structure-from-Motion (SfM) software like COLMAP to build a 3D model. However, if your images contain visually similar but distinct elements (like the opposite sides of a symmetric building), SfM can get confused and produce a faulty 3D model. This tool helps by analyzing pairs of images and identifying these 'doppelganger' pairs, feeding the refined list to COLMAP for a more accurate 3D reconstruction. This is for 3D modelers, architects, or anyone needing precise 3D reconstructions from photographs.
198 stars. No commits in the last 6 months.
Use this if you are creating 3D models from many photos of highly repetitive or symmetric structures and are encountering errors or inaccuracies in your reconstructions.
Not ideal if your image sets rarely contain highly similar but distinct visual elements, or if you are not performing 3D reconstruction using SfM.
Stars
198
Forks
28
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Mar 01, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/RuojinCai/doppelgangers"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
anchen1011/toflow
TOFlow: Video Enhancement with Task-Oriented Flow
NVlabs/nvdiffrec
Official code for the CVPR 2022 (oral) paper "Extracting Triangular 3D Models, Materials, and...
cvg/GlueStick
Joint Deep Matcher for Points and Lines 🖼️💥🖼️ (ICCV 2023)
microsoft/SpareNet
Style-based Point Generator with Adversarial Rendering for Point Cloud Completion (CVPR 2021)
fraunhoferhhi/neural-deferred-shading
Multi-View Mesh Reconstruction with Neural Deferred Shading (CVPR 2022)