DC-Gaussian: Improving 3D Gaussian Splatting for Reflective Dash Cam Videos

TL;DR

DC-Gaussian is a novel neural rendering method tailored for dash cam videos, effectively handling reflections and occlusions to produce high-quality novel views and scene reconstructions under diverse lighting conditions.

Contribution

It introduces adaptive image decomposition, illumination-aware obstruction modeling, and geometry-guided enhancement to improve neural rendering for obstructed dash cam videos.

Findings

Achieves state-of-the-art novel view synthesis performance.

Effectively models reflections and occlusions in dash cam videos.

Accurately reconstructs scenes while removing obstructions.

Abstract

We present DC-Gaussian, a new method for generating novel views from in-vehicle dash cam videos. While neural rendering techniques have made significant strides in driving scenarios, existing methods are primarily designed for videos collected by autonomous vehicles. However, these videos are limited in both quantity and diversity compared to dash cam videos, which are more widely used across various types of vehicles and capture a broader range of scenarios. Dash cam videos often suffer from severe obstructions such as reflections and occlusions on the windshields, which significantly impede the application of neural rendering techniques. To address this challenge, we develop DC-Gaussian based on the recent real-time neural rendering technique 3D Gaussian Splatting (3DGS). Our approach includes an adaptive image decomposition module to model reflections and occlusions in a unified manner. Additionally, we introduce illumination-aware obstruction modeling to manage reflections and occlusions under varying lighting conditions. Lastly, we employ a geometry-guided Gaussian enhancement strategy to improve rendering details by incorporating additional geometry priors. Experiments on self-captured and public dash cam videos show that our method not only achieves state-of-the-art performance in novel view synthesis, but also accurately reconstructing captured scenes getting rid of obstructions. See the project page for code, data: https://linhanwang.github.io/dcgaussian/.