Research map / 3D vision / 3D generation
PanopticNeRF-360: Panoramic 3D-to-2D Label Transfer in Urban Scenes
Xiao Yun Fu, Shangzhan Zhang, Tianrun Chen, Yichong Lu, Xiaowei Zhou, Andreas Geiger, Yiyi Liao
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 PanopticNeRF-360 instead of 3D-2D CRF | label transfer · KITTI-360 fisheye | mIoU: 64.4 -> 76.3 | Improved |
| use PanopticNeRF-360 instead of 3D-2D CRF | label transfer · KITTI-360 forward-facing | mIoU: 79.5 -> 81.9 | Improved |
| add label-guided geometry loss on fixed semantic field (ablation) | PanopticNeRF-360 · KITTI-360 one scene | depth RMSE: 16.90 -> 7.27 | Improved |
| add fisheye semantic predictions as input (ablation) | PanopticNeRF-360 · KITTI-360 one scene | mIoU*: 74.8 -> 76.1 | Improved |
| add 2D loss on learned semantic field (ablation) | PanopticNeRF-360 · KITTI-360 one scene | mIoU*: 74.6 -> 76.1 | Improved |
| add per-point 3D semantic loss (ablation) | PanopticNeRF-360 · KITTI-360 one scene | mIoU*: 75.0 -> 76.1 | Improved |
| add weak stereo depth loss (ablation) | PanopticNeRF-360 · KITTI-360 one scene | mIoU*: 74.4 -> 76.1 | Improved |
| add ray masking (ablation) | PanopticNeRF-360 · KITTI-360 one scene | mIoU*: 75.1 -> 76.1 | Improved |
| use bounding-primitive-guided sampling instead of uniform sampling (ablation) | PanopticNeRF-360 · KITTI-360 one scene | mIoU*: 67.3 -> 76.1 | Improved |
| add class-frequency weight w(k) (ablation) | PanopticNeRF-360 · KITTI-360 one scene | mIoU*: 75.6 -> 76.1 | Mixed |
| use hybrid MLP+hash grid instead of pure MLP | scene representation · KITTI-360 forward-facing (4 scenes) | PSNR: 23.78 -> 27.25 | Mixed |
| use hybrid MLP+hash grid instead of iNGP | scene representation · KITTI-360 forward-facing (4 scenes) | mIoU: 79.0 -> 80.9 | Improved |
Closest papers by meaning
- PanDA: Unsupervised Domain Adaptation for Multimodal 3D Panoptic Segmentation in Autonomous Driving (CVPR 2026 · similarity 0.88)
- SGFormer: Semantic-Geometry Fusion Transformer for Multi-modal 3D Panoptic Segmentation (AAAI 2025 · similarity 0.88)
- Panoptic Neural Fields: A Semantic Object-Aware Neural Scene Representation (CVPR 2022 · similarity 0.88)
- MMScan: A Multi-Modal 3D Scene Dataset with Hierarchical Grounded Language Annotations (NeurIPS 2024 · similarity 0.86)
- Building a Strong Pre-Training Baseline for Universal 3D Large-Scale Perception (CVPR 2024 · similarity 0.86)
- Semi-supervised 3D Semantic Scene Completion with 2D Vision Foundation Model Guidance (AAAI 2025 · similarity 0.86)
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.
