nianticspatial / ace-g
[ICCV 2025] ACE-G is an architecture and pre-training scheme to improve generalization for scene coordinate regression-based visual relocalization.
Why this job: topics: computer-vision, pose-estimation, deep-learning, machine-learning
What it does
This repository contains the code for "ACE-G: Improving Scene Coordinate Regression Through Query Pre-Training". For illustrations of the method and additional visualizations check out the project page and the paper. We provide the code to find the map code for novel scenes, relocalize images within, visualize the results, and reproduce the results from the paper. Currently, there are no plans to release the code for pre-training, but we do provide the pre-trained model used in the paper.
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Inactive
- Last commit 8 months ago
- Stars
- 110
- On GitHub
- Licence
- Custom
- Read before business use
- Downloads for
- None listed
- No release files
- Latest version
- None yet
- No releases
- Built with
- Python
- 69% of the code
Is it looked after?
- Contributors
- Not stated
- Last 90 days
- Issues
- 100% closed
- 0 still open
- Pull requests
- 0 merged
- 0 waiting
- Releases
- None
- Never published
- No contributing guide
- No security policy
- No code of conduct
More: languages, topics and project record
Languages
- Python 69%
- C++ 31%
- Shell 1%
Topics
- camera-relocalization
- computer-vision
- deep-learning
- iccv
- machine-learning
- pose-estimation
- scene-coordinate-regression
- visual-relocalization
Project record
- JobAI libraries and models
- StatusListed
- Commits5
- Forks7
- Watching2
- Stars in 28 days+3
- CreatedOct 9, 2025
- First seen by usNov 7, 2026
- CheckedNov 8, 2026