LidarPainter: One-Step Away From Any Lidar View To Novel Guidance
This addresses reconstruction quality issues for digital twin systems and autonomous driving simulation, offering a novel method for real-time, high-fidelity lane shifts.
The paper tackles the problem of degraded dynamic driving scene reconstruction when views deviate from input trajectories, proposing LidarPainter, a one-step diffusion model that recovers consistent views from sparse LiDAR data and corrupted renderings in real-time, achieving 7x faster speed and using one-fifth the GPU memory compared to state-of-the-art methods.
Dynamic driving scene reconstruction is of great importance in fields like digital twin system and autonomous driving simulation. However, unacceptable degradation occurs when the view deviates from the input trajectory, leading to corrupted background and vehicle models. To improve reconstruction quality on novel trajectory, existing methods are subject to various limitations including inconsistency, deformation, and time consumption. This paper proposes LidarPainter, a one-step diffusion model that recovers consistent driving views from sparse LiDAR condition and artifact-corrupted renderings in real-time, enabling high-fidelity lane shifts in driving scene reconstruction. Extensive experiments show that LidarPainter outperforms state-of-the-art methods in speed, quality and resource efficiency, specifically 7 x faster than StreetCrafter with only one fifth of GPU memory required. LidarPainter also supports stylized generation using text prompts such as "foggy" and "night", allowing for a diverse expansion of the existing asset library.