Research map / 3D vision / Face and shape
EMIE-MAP: Large-Scale Road Surface Reconstruction Based on Explicit Mesh and Implicit Encoding
Wenhua Wu, Qi Wang, Guangming Wang, Junping Wang, Tiankun Zhao, Yang Liu, Dongchao Gao, Zhe Liu, Hesheng Wang
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 explicit mesh with implicit color encoding and elevation residual MLP instead of RoMe | RoMe · Custom dataset Night | PSNR: 0.99 -> 22.12 | Improved |
| use EMIE-MAP instead of RoMe | RoMe · Custom dataset Ramp | Elev-error (cm): 496.24 -> 93.12 | Improved |
| use EMIE-MAP instead of RoMe | RoMe · KITTI-odometry sequence-09 | Elev-error (cm): 107.98 -> 28.12 | Improved |
| add Lidar ground-height supervision | EMIE-MAP · Custom dataset City street | Elev-error (cm): 17.08 -> 2.35 | Improved |
| add elevation residual MLP (ablation) | EMIE-MAP · Custom dataset City street | Elev-error (cm): 12.78 -> 2.35 | Improved |
| add per-camera RGB MLPs on shared implicit color encoding (ablation) | EMIE-MAP · Custom dataset City street | PSNR: 15.60 -> 26.75 | Improved |
| use different MLPs per camera instead of a single MLP with embeddings (ablation) | EMIE-MAP · Custom dataset City street | PSNR: 25.83 -> 26.75 | Improved |
| add semantic branch (ablation) | EMIE-MAP · Custom dataset City street | PSNR: 26.37 -> 26.75 | Improved |
| add RGB MLPs on elevation accuracy (ablation) | EMIE-MAP · Custom dataset City street | mIoU (%): 95.16 -> 95.27 | No clear effect |
Closest papers by meaning
- ViiNeuS: Volumetric Initialization for Implicit Neural Surface Reconstruction of Urban Scenes with Limited Image Overlap (CVPR 2025 · similarity 0.87)
- Swift-Mapping: Online Neural Implicit Dense Mapping in Urban Scenes (AAAI 2024 · similarity 0.86)
- BEV-GS: Feed-forward Gaussian Splatting in Bird’s-Eye-View for Road Reconstruction (ECCV 2026 · similarity 0.85)
- READ: Large-Scale Neural Scene Rendering for Autonomous Driving (AAAI 2023 · similarity 0.85)
- LSD-3D: Large-Scale 3D Driving Scene Generation with Geometry Grounding (AAAI 2026 · similarity 0.85)
- LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving (ICLR 2024 · 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.
