3D GAUSSIAN SPLATTING
This research project aims to gain insight on the topic of 3D Gaussian Splatting, understand its pipeline, and learn about its applications.
Learnings
The main premise of 3D Gaussian Splatting is to recreate photorealistic depictions of physical scenes by encoding 2D images to create novel viewpoints. First, a point cloud is generated, using Structure-for-Motion, for each image in the dataset. For each point, a 3D gaussian is initialized at the center with the attributes opacity, covariance, mean, and an RGB color parameter. With each iteration of rendering, the gaussians are optimized to represent the scene geometry.
This new field of research is extremely interesting because being able to render something as high resolution and as quickly assists researchers in studying areas impossible for humans to sustain life for an extended period of time.
DEMO
I performed a 3D Gaussian Splat of a classroom following this repo. This scene is trained on 84 frames and rendered with 7,000 point clouds. Some interesting aspects to note are the rendered details on features I didn't mean to capture in the video. For example, the high resolution writing on the chalkboard in the background and the captured fur detailing of the stuffed animal!