AIBluefisher/DAGSfM
A distributed, graph-based structure-from-motion library for large-scale 3D reconstruction from 2D image sequences.
Not currently ranked — collecting fresh signals.
star history
This library implements graph-based parallel algorithms for large-scale structure-from-motion, enabling distributed 3D reconstruction from collections of images. It uses community detection on feature correspondence graphs to partition reconstruction tasks across workers, then merges results. The implementation relies on traditional photogrammetric techniques including bundle adjustment via ceres-solver, igraph for graph analysis, and RPC-based distributed communication, rather than deep learning approaches.
Frequently asked
- What is AIBluefisher/DAGSfM?
- A distributed, graph-based structure-from-motion library for large-scale 3D reconstruction from 2D image sequences.
- Is DAGSfM open source?
- Yes — AIBluefisher/DAGSfM is open source, released under the BSD-3-Clause license.
- What language is DAGSfM written in?
- AIBluefisher/DAGSfM is primarily written in C.
- How popular is DAGSfM?
- AIBluefisher/DAGSfM has 422 stars on GitHub.
- Where can I find DAGSfM?
- AIBluefisher/DAGSfM is on GitHub at https://github.com/AIBluefisher/DAGSfM.