Research map / 3D vision / Human pose estimation
Expert Sample Consensus Applied to Camera Re-Localization
Eric Brachmann, Carsten Rother
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 |
|---|---|---|---|
| use ESAC instead of Expert Selection | expert ensemble · Toy problem | parameter accuracy: +3.9% | Improved |
| use ESAC instead of Expert Selection | expert ensemble · Toy problem | model type classification accuracy: 11.5% | Improved |
| limit ESAC to top 2 experts per test image vs Expert Selection (ablation) | ESAC · 19Scenes | pose accuracy: +19.7% | Improved |
| use ESAC with distributed hypotheses instead of a single monolithic DSAC++ | DSAC++ · 19Scenes | pose accuracy 5cm/5deg: 53.3% -> 88.1% | Improved |
| end-to-end training of gating network, experts or all (ablation) | ESAC · 19Scenes | pose accuracy | Improved |
| end-to-end training evaluated at stricter 2cm/2deg threshold (ablation) | ESAC · 19Scenes | pose accuracy: >10% | Improved |
| replace VGG with ResNet experts | DSAC++ architecture · 7Scenes/12Scenes | pose accuracy and memory: 210MB -> 28MB | No clear effect |
| increase number of experts (ablation) | ESAC · Dubrovnik/Aachen Day | pose accuracy | No clear effect |
Closest papers by meaning
- Unsupervised Learning of Consensus Maximization for 3D Vision Problems (CVPR 2019 · similarity 0.86)
- On the Instability of Relative Pose Estimation and RANSAC's Role (CVPR 2022 · similarity 0.86)
- BANSAC: A Dynamic BAyesian Network for Adaptive SAmple Consensus (ICCV 2023 · similarity 0.86)
- PARSAC: Accelerating Robust Multi-Model Fitting with Parallel Sample Consensus (AAAI 2024 · similarity 0.85)
- Condition Numbers in Multiview Geometry, Instability in Relative Pose Estimation, and RANSAC (TPAMI 2025 · similarity 0.85)
- CONSAC: Robust Multi-Model Fitting by Conditional Sample Consensus (CVPR 2020 · 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.
