Gaurav Sharma, Ravi Kothari, Josef Schmid
The paper presents a self-supervised NeRF framework for autonomous driving that efficiently learns from LiDAR and camera data without 3D labels, achieving superior performance on the KITTI-360 dataset.
This research introduces a new framework for creating 3D models from both LiDAR and camera data, which is important for self-driving cars. The framework, called Novel View Synthesis Framework (NVSF), doesn't need labeled 3D data, making it easier to use in real-world applications. It uses smart techniques to focus on important parts of images and preserve detailed features from LiDAR data. Tests show that it performs better than existing methods on a standard dataset used for evaluating autonomous driving systems.
