Research map / Image generation / Autoregressive and latent generation
Latent Semantic Consensus for Deterministic Geometric Model Fitting
Guobao Xiao, Jun Yu, Jiayi Ma, Deng-Ping Fan, Ling Shao
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 LSC-SA latent-semantic sampling instead of RANSAC sampling | MSSC-style model selection · Hopkins155 | mean segmentation error (%): 0.83 -> 0.71 | Improved |
| sample from affine, homography and fundamental models (Subset) instead of homography only | LSC · Hopkins155 | mean segmentation error (%): 0.71 -> 0.31 | Improved |
| use LSC instead of Subset for speed | Subset · Hopkins155 | CPU time (s): 2621.80 -> 1508.19 | Improved |
| use LSC instead of RANSAC | RANSAC · KITTI sequence 02 | translation error (m): 4.01 -> 2.24 | Improved |
| use LSC instead of RANSAC | RANSAC · KITTI sequence 00 | CPU time (s): 1.07 -> 0.33 | Improved |
| use LSC (high-quality hypotheses) instead of CLSA selection on poor hypotheses | CLSA · 4 circles synthetic | segmentation error (%): 38.55 -> 1.25 | Improved |
| use LSC instead of CLSA | CLSA · 4 lines synthetic | segmentation error (%): 2.28 -> 2.63 | Hurt |
| use LSC instead of RansaCov | RansaCov · 5 circles synthetic | segmentation error (%): 0.23 -> 0.33 | Hurt |
| set threshold psi to 0.05 for homography estimation (ablation) | LSC · AdelaideRMF homography | segmentation error | Improved |
| set threshold psi to 0.01 for fundamental matrix estimation (ablation) | LSC · AdelaideRMF fundamental matrix | segmentation error | Improved |
| set threshold beta to 0.8 (ablation) | LSC · AdelaideRMF | segmentation error | Improved |
| vary psi from 0.01 to 0.10 (ablation) | LSC · MS-COCO-F homography | segmentation error | No clear effect |
Closest papers by meaning
- Finding Geometric Models by Clustering in the Consensus Space (CVPR 2023 · similarity 0.85)
- PARSAC: Accelerating Robust Multi-Model Fitting with Parallel Sample Consensus (AAAI 2024 · similarity 0.84)
- Relaxed Class-consensus Consistency for Semi-supervised Semantic Segmentation (AAAI 2025 · similarity 0.83)
- Robust Multi-view Clustering against Imperfect Information (NeurIPS 2026 · similarity 0.83)
- Improved Training Technique for Latent Consistency Models (ICLR 2025 · similarity 0.83)
- Large Spatial Model: End-to-end Unposed Images to Semantic 3D (NeurIPS 2024 · similarity 0.83)
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.
