Research map / Point clouds / Point cloud classification
Detecting Vanishing Points Using Global Image Context in a Non-Manhattan World
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 horizon-first vanishing-point detection with CNN global context | Horizon-line detection · YUD | AUC: 94.51% -> 94.78% | Helped |
| Use horizon-first vanishing-point detection with CNN global context | Horizon-line detection · ECD | AUC: 89.20% -> 90.80% | Helped |
| Use horizon-first vanishing-point detection with CNN global context | Horizon-line detection · HLW test | AUC: 52.59% -> 58.24% | Helped |
| Use the proposed method instead of Lezama et al. | Horizon-line detection · YUD | seconds per image: approximately 30 seconds -> 1 second | Helped |
| Use the proposed method instead of Lezama et al. | Horizon-line detection · ECD | seconds per image: approximately 60 seconds -> 3 seconds | Helped |
| Add CNN global context to horizon-first vanishing-point detection (ablation) | Horizon-first vanishing-point detection · YUD | AUC: 93.87% -> 94.78% | Helped |
| Add CNN global context to horizon-first vanishing-point detection (ablation) | Horizon-first vanishing-point detection · ECD | AUC: 87.94% -> 90.80% | Helped |
| Add CNN global context to horizon-first vanishing-point detection (ablation) | Horizon-first vanishing-point detection · HLW test | AUC: 53.04% -> 58.24% | Helped |
| Replace GIST random-forest context with CNN context (ablation) | Full horizon-first vanishing-point detection · YUD | AUC: 94.66% -> 94.78% | Helped |
| Replace GIST random-forest context with CNN context (ablation) | Full horizon-first vanishing-point detection · ECD | AUC: 87.60% -> 90.80% | Helped |
| Replace GIST random-forest context with CNN context (ablation) | Full horizon-first vanishing-point detection · HLW test | AUC: 54.95% -> 58.24% | Helped |
| Add vanishing-point detection to CNN horizon prediction (ablation) | CNN global-context horizon prediction · YUD | AUC: 73.67% -> 94.78% | Helped |
Closest papers by meaning
- End-to-End Real-Time Vanishing Point Detection with Transformer (AAAI 2024 · similarity 0.89)
- Convex Relaxation for Robust Vanishing Point Estimation in Manhattan World (CVPR 2025 · similarity 0.89)
- VPDETR: End-to-End Vanishing Point DEtection TRansformers (AAAI 2024 · similarity 0.88)
- GeoComplete: Geometry-Aware Diffusion for Reference-Driven Image Completion (NeurIPS 2025 · similarity 0.84)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.83)
- DeCoTR: Enhancing Depth Completion with 2D and 3D Attentions (CVPR 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.