Research map / 3D vision / 3D generation
Sensor2Sensor: Cross-Embodiment Sensor Conversion for Autonomous Driving
Jiahao Wang, Bo Sun, Yijing Bai, Vincent Casser, Songyou Peng, Zehao Zhu, Meng-Li Shih, Xander Masotto, Shih-Yang Su, Kanaad Parvate, Tiancheng Ge, Linn Bieske and 3 more
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 |
|---|---|---|---|
| view concatenation of dashcam instead of channel concatenation (ablation) | Sensor2Sensor · Fixed-Camera-to-AV images | FID: 6.88 -> 6.47 | Improved |
| view concatenation instead of channel concatenation (image-only) (ablation) | CAT3D · Fixed-Camera-to-AV images | FID: 6.63 -> 6.20 | Improved |
| view concatenation (ablation) | Sensor2Sensor · Fixed-Camera-to-AV video | FVD: 293.73 -> 278.12 | Improved |
| view concatenation (ablation) | Sensor2Sensor · Fixed-Camera-to-AV images | LPIPS: 0.346 -> 0.316 | Improved |
| add joint LiDAR training to view-concatenation model (ablation) | Sensor2Sensor · Fixed-Camera-to-AV images | FID: 6.20 -> 6.47 | Hurt |
| add joint LiDAR training to view-concatenation model (ablation) | Sensor2Sensor · Fixed-Camera-to-AV images | LPIPS: 0.307 -> 0.316 | Hurt |
| add DAgger finetuning (ablation) | Sensor2Sensor video · Fixed-Camera-to-AV front-view video | FVD: 288.90 -> 278.12 | Improved |
| add DAgger finetuning (ablation) | Sensor2Sensor video · Fixed-Camera-to-AV front-view video | FID: 24.65 -> 21.54 | Improved |
| Sensor2Sensor instead of X-Drive for LiDAR | Sensor2Sensor · LiDAR generation | Chamfer Distance: 10.02 -> 8.68 | Improved |
| Sensor2Sensor instead of X-Drive for images | Sensor2Sensor · Fixed-Camera-to-AV images | FID: 8.30 -> 6.47 | Improved |
| Sensor2Sensor vs feedforward reconstruction (pi^3) | Sensor2Sensor · Fixed-Camera-to-AV video | FVD: 2007.35 -> 278.12 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- VGGT — VGGT: Visual Geometry Grounded Transformer (CVPR 2025) · better on 8 of 8 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting (NeurIPS 2021 · similarity 0.87)
- Glad: A Streaming Scene Generator for Autonomous Driving (ICLR 2025 · similarity 0.87)
- Panacea: Panoramic and Controllable Video Generation for Autonomous Driving (CVPR 2024 · similarity 0.86)
- NeuRAD: Neural Rendering for Autonomous Driving (CVPR 2024 · similarity 0.86)
- Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis (NeurIPS 2026 · similarity 0.85)
- The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset (NeurIPS 2026 · 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.
