Research map / Point clouds / Registration and matching
Globally Optimal and Efficient Vanishing Point Estimation in Atlanta World
Haoang Li, Pyojin Kim, Ji Zhao, Kyungdon Joo, Zhipeng Cai, Zhe Liu, Yun-Hui Liu
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 pure BnB over all VPs with BnB plus mine-and-stab (OnI) | BnB · Our dataset, manual lines (117 lines, 4 VPs) | run time (sec): 12.85 -> 3.06 | Improved |
| replace pure BnB with OnI | BnB · NYU, manual lines (79 lines, 4 VPs) | run time (sec): 9.92 -> 2.15 | Improved |
| replace pure BnB with OnI | BnB · Our dataset, LSD lines (84 lines, 5 VPs) | run time (sec): 11.09 -> 2.48 | Improved |
| replace pure BnB with OnI | BnB · NYU, LSD lines (86 lines, 4 VPs) | run time (sec): 13.10 -> 2.73 | Improved |
| replace pure BnB with OnI | BnB · NYU, LSD lines (86 lines, 4 VPs) | consistency error (pixel): 0.60 -> 0.62 | No clear effect |
| replace pure BnB with OnI | BnB · Our dataset, manual lines (117 lines, 4 VPs) | consistency error (pixel): 0.44 -> 0.45 | No clear effect |
| use ground-truth vertical direction (Inner) instead of searching it with BnB (OnI) (ablation) | OnI · Our dataset, manual lines (117 lines, 4 VPs) | run time (sec): 3.06 -> 0.04 | Improved |
| use ground-truth vertical direction (Inner) instead of searching it (ablation) | OnI · NYU, LSD lines (86 lines, 4 VPs) | consistency error (pixel): 0.62 -> 0.55 | Improved |
| use ground-truth vertical direction (Inner) instead of searching it (ablation) | OnI · Our dataset, LSD lines (84 lines, 5 VPs) | consistency error (pixel): 0.75 -> 0.61 | Improved |
| replace Hough with Inner | Hough · NYU, LSD lines (86 lines, 4 VPs) | consistency error (pixel): 9.03 -> 0.55 | Improved |
| replace T-Linkage with OnI | T-Linkage · Our dataset, LSD lines (84 lines, 5 VPs) | consistency error (pixel): 1.46 -> 0.75 | Improved |
Closest papers by meaning
- Quasi-Globally Optimal and Near/True Real-Time Vanishing Point Estimation in Manhattan World (TPAMI 2020 · similarity 0.92)
- Quasi-Globally Optimal and Efficient Vanishing Point Estimation in Manhattan World (ICCV 2019 · similarity 0.92)
- Convex Relaxation for Robust Vanishing Point Estimation in Manhattan World (CVPR 2025 · similarity 0.91)
- Globally Optimal Inlier Set Maximization for Atlanta World Understanding (TPAMI 2019 · similarity 0.89)
- VPDETR: End-to-End Vanishing Point DEtection TRansformers (AAAI 2024 · similarity 0.86)
- End-to-End Real-Time Vanishing Point Detection with Transformer (AAAI 2024 · similarity 0.84)
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.
