Research map / Detection and segmentation / CNN features
Slanted Stixels: A Way to Represent Steep Streets
Daniel Hernandez-Juarez, Lukas Schneider, Pau Cebrian, Antonio Espinosa, David Vázquez, Antonio Manuel López, Uwe Franke, Marc Pollefeys, Juan Carlos Moure
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 flat-road depth model with slanted Stixel depth model | Semantic Stixels · SYNTHIA-SF (non-flat roads) | Disparity error (%): 29.33 -> 12.99 | Improved |
| replace flat-road depth model with slanted Stixel depth model | Semantic Stixels · SYNTHIA-SF (non-flat roads) | IoU (%): 46.0 -> 48.5 | Improved |
| replace flat-road depth model with slanted Stixel depth model | Semantic Stixels · Ladicky (flat roads) | Disparity error (%): 17.38 -> 16.84 | Improved |
| replace flat-road depth model with slanted Stixel depth model | Semantic Stixels · KITTI 15 (flat roads) | Disparity error (%): 11.05 -> 11.21 | No clear effect |
| replace flat-road depth model with slanted Stixel depth model | Semantic Stixels · Ladicky (flat roads) | IoU (%): 66.2 -> 66.1 | No clear effect |
| replace flat-road depth model with slanted Stixel depth model | Semantic Stixels · Cityscapes | IoU (%): 65.4 -> 65.8 | Improved |
| add Fast over-segmentation (Time Series) to Slanted Stixels (ablation) | Slanted Stixels · SYNTHIA-SF | Disparity error (%): 12.99 -> 14.20 | Hurt |
| add Fast over-segmentation (FCN) to Slanted Stixels (ablation) | Slanted Stixels · SYNTHIA-SF | Disparity error (%): 12.99 -> 14.19 | Hurt |
| add Fast over-segmentation (FCN) to Slanted Stixels (ablation) | Slanted Stixels · Ladicky | Disparity error (%): 16.84 -> 16.84 | No clear effect |
| use FCN-based over-segmentation instead of Time Series (ablation) | Slanted Stixels · SYNTHIA-SF | IoU (%): 48.0 -> 48.6 | Improved |
| add over-segmentation (FCN) for faster inference (ablation) | Slanted Stixels · Per-stage run-time (Stixel stage) | Frame-rate (Hz): 61 -> 130 | Improved |
| add over-segmentation (Time Series) (ablation) | Slanted Stixels · Per-stage run-time (Stixel stage) | Frame-rate (Hz): 61 -> 116 | Improved |
Closest papers by meaning
- Scene Representation Transformer: Geometry-Free Novel View Synthesis Through Set-Latent Scene Representations (CVPR 2022 · similarity 0.85)
- Learning to Look around Objects for Top-View Representations of Outdoor Scenes (ECCV 2018 · similarity 0.85)
- Sketch and Refine: Towards Fast and Accurate Lane Detection (AAAI 2024 · similarity 0.84)
- GSENet:Global Semantic Enhancement Network for Lane Detection (AAAI 2024 · similarity 0.84)
- ReManNet: A Riemannian Manifold Network for Monocular 3D Lane Detection (CVPR 2026 · similarity 0.84)
- 3D StreetUnveiler with Semantic-aware 2DGS - a simple baseline (ICLR 2025 · 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.
