Research map / 3D vision / Depth and stereo
Adversarial Exploitation of Data Diversity Improves Visual Localization
Sihang Li, Siqi Tan, Bowen Chang, Jing Zhang, Chen Feng, Yiming Li
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 VGG16 with Efficient-B0 feature extractor (ablation) | RAP baseline · Cambridge Shop | translation error (cm): 174 -> 103 | Improved |
| add pose augmentation from 3DGS (ablation) | RAP · Cambridge Shop | translation error (cm): 103 -> 75 | Improved |
| add appearance augmentation (ablation) | RAP · Cambridge Shop | translation error (cm): 75 -> 60 | Improved |
| add ConvNet decoder between features and regression (ablation) | RAP · Cambridge Shop | translation error (cm): 60 -> 52 | Improved |
| use Pose Transformer instead of ConvNet decoder (ablation) | RAP · Cambridge Shop | translation error (cm): 52 -> 40 | Improved |
| add adversarial discriminator (ablation) | RAP · Cambridge Shop | translation error (cm): 40 -> 33 | Improved |
| use our pose augmentation instead of LENS policy (ablation) | RAP · Cambridge Landmarks | translation error (cm): 100 -> 56 | Improved |
| add appearance augmentation (ablation) | RAP · Aachen Day-Night | rotation error (deg): 75.99 -> 13.70 | Improved |
| add augmentation to real training data (100% poses) (ablation) | RAP · Cambridge Shop | median translation error (cm): 98 -> 33 | Improved |
| train on 20% real data with augmentation instead of 100% without (ablation) | RAP · Cambridge Shop | median translation error (cm): 98 -> 87 | Improved |
| replace NeRF with 3DGS naively in DFNet (ablation) | DFNet · Cambridge Landmarks | rotation error (deg): 2.90 -> 4.75 | Hurt |
| enlarge void zone around test camera to 30 cm / 1.5 deg (ablation) | RAP · Cambridge Shop | median translation error (cm): 33 -> 40 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- GLACE — GLACE: Global Local Accelerated Coordinate Encoding (CVPR 2024) · better on 1 of 2 shared results
- DFNet — DFNet: Enhance Absolute Pose Regression with Direct Feature Matching (ECCV 2022) · better on 2 of 2 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
- HyperPose: Hypernetwork-Infused Camera Pose Localization and an Extended Cambridge Landmarks Dataset (CVPR 2025 · similarity 0.87)
- ConDo: Continual Domain Expansion for Absolute Pose Regression (AAAI 2025 · similarity 0.86)
- Weatherproofing Retrieval for Localization with Generative AI and Geometric Consistency (ICLR 2024 · similarity 0.86)
- R-SCoRe: Revisiting Scene Coordinate Regression for Robust Large-Scale Visual Localization (CVPR 2025 · similarity 0.86)
- Enhancing Visual Localization with Cross-Domain Image Generation (ICML 2025 · similarity 0.86)
- Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localization (CVPR 2025 · similarity 0.85)
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.
