Research map / 3D vision / Gaussian splatting
DGGT: Feedforward 4D Reconstruction of Dynamic Driving Scenes using Unposed Images
Xiaoxue Chen, Ziyi Xiong, Yuantao Chen, Gen Li, Nan Wang, Hongcheng Luo, Long Chen, Haiyang Sun, Bing Wang, Guang Chen, Hongyang Li, Ya-Qin Zhang and 2 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 |
|---|---|---|---|
| add lifespan parameter (ablation) | DGGT · Waymo NVS | PSNR: 24.21 -> 27.41 | Improved |
| add single-step diffusion refinement (ablation) | DGGT · Waymo NVS | PSNR: 27.32 -> 27.41 | Improved |
| add single-step diffusion refinement (ablation) | DGGT · Waymo NVS | LPIPS: 0.108 -> 0.109 | No clear effect |
| increase input views from 4 to 16 | STORM · Waymo NVS | PSNR: 26.05 -> 22.98 | Hurt |
| increase input views from 4 to 16 | DGGT · Waymo NVS | PSNR: 27.41 -> 28.14 | Improved |
| predict pose as output (pose-free) instead of input | DGGT · nuScenes zero-shot | PSNR: 19.75 -> 25.31 | Improved |
| train on target dataset after Waymo | DGGT · nuScenes | PSNR: 25.31 -> 26.63 | Improved |
| model full 3D motion trajectories with motion head instead of velocities | DGGT · Waymo Scene Flow | EPE3D: 0.276 -> 0.183 | Improved |
| use DGGT instead of STORM | feedforward reconstruction · Waymo NVS | inference time: 0.18s -> 0.39s | Hurt |
Closest papers by meaning
- DrivingForward: Feed-forward 3D Gaussian Splatting for Driving Scene Reconstruction from Flexible Surround-view Input (AAAI 2025 · similarity 0.91)
- One World, Dual Timeline: Decoupled Spatio-Temporal Gaussian Scene Graph for 4D Cooperative Driving Reconstruction (NeurIPS 2026 · similarity 0.90)
- DrivingGaussian: Composite Gaussian Splatting for Surrounding Dynamic Autonomous Driving Scenes (CVPR 2024 · similarity 0.90)
- FRUC: Feedforward Dynamic Scene Reconstruction from Uncalibrated Collaborative Driving Views (NeurIPS 2026 · similarity 0.90)
- DrivingRecon: Large 4D Gaussian Reconstruction Model For Autonomous Driving (NeurIPS 2025 · similarity 0.89)
- Flux4D: Flow-based Unsupervised 4D Reconstruction (NeurIPS 2025 · similarity 0.89)
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.
