Research map / 3D vision / Depth and stereo
ACE-G: Improving Generalization of Scene Coordinate Regression Through Query Pre-Training
Leonard Bruns, Axel Barroso-Laguna, Tommaso Cavallari, Aron Monszpart, Sowmya Munukutla, Victor Adrian Prisacariu, Eric Brachmann
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace ACE encoder with DINOv2 features (ablation) | ACE · Indoor-6 | % localized: 70 -> 91 | Improved |
| add query pre-training of scene-agnostic regressor (ACE-G vs DINO-ACE) (ablation) | DINO-ACE · Indoor-6 | % localized: 91 -> 95 | Improved |
| add query pre-training (ACE-G vs DINO-ACE) (ablation) | DINO-ACE · RIO10 | median translation error (cm): 83.8 -> 41.1 | Improved |
| add query pre-training (ACE-G vs DINO-ACE) (ablation) | DINO-ACE · Cambridge Landmarks | median translation error (cm): 43.9 -> 35.1 | Improved |
| add query pre-training with ACE encoder (ablation) | ACE · RIO10 | median position error (cm): 358.4 -> 144.5 | Improved |
| add query pre-training with ACE encoder (ablation) | ACE · 7Scenes | median position error (cm): 1.1 -> 1.3 | No clear effect |
| replace ACE encoder with DINOv2 encoder (ablation) | ACE-G · 7Scenes (static) | median position error (cm): 1.3 -> 4.6 | Hurt |
| add query iterations during pre-training (ablation) | ACE-G · validation splits | accuracy (20deg,20cm) | Improved |
| add uncertainty-based quantile prefiltering of scene coordinates (ablation) | ACE-G · registration | pose accuracy | Improved |
| use ACE-G instead of ACE at fine threshold | ACE · Cambridge (10deg/10cm) | % localized: 26.3 -> 17.9 | Hurt |
| use ACE-G (5 min) instead of ACE | ACE · Indoor-6 (5deg/5cm) | % localized: 33.5 -> 54.7 | Improved |
| fine-tune on indoor datasets only (ablation) | ACE-G · SN, SN++, ARK vs MF, BMVG, WR | accuracy (20deg,20cm): +1.9 / +2.6 / +1.2 on SN, SN++, ARK; -3.4 / -7.5 / -9.2 on MF, BMVG, WR | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- GLACE — GLACE: Global Local Accelerated Coordinate Encoding (CVPR 2024) · better on 2 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localization (CVPR 2025 · similarity 0.91)
- GLACE: Global Local Accelerated Coordinate Encoding (CVPR 2024 · similarity 0.90)
- Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localization (CVPR 2025 · similarity 0.89)
- Map-Relative Pose Regression for Visual Re-Localization (CVPR 2024 · similarity 0.88)
- Where Am I and What Will I See: An Auto-Regressive Model for Spatial Localization and View Prediction (ICLR 2025 · similarity 0.86)
- Scene-agnostic Pose Regression for Visual Localization (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
