TC-GS: Tri-plane based compression for 3D Gaussian Splatting
This addresses the memory cost issue for practical applications of 3DGS, representing an incremental improvement in compression techniques.
The paper tackles the problem of compressing the substantial data volume of 3D Gaussian Splatting (3DGS) for novel view synthesis by proposing a tri-plane based compression method, achieving results comparable to or surpassing state-of-the-art compression work across multiple datasets.
Recently, 3D Gaussian Splatting (3DGS) has emerged as a prominent framework for novel view synthesis, providing high fidelity and rapid rendering speed. However, the substantial data volume of 3DGS and its attributes impede its practical utility, requiring compression techniques for reducing memory cost. Nevertheless, the unorganized shape of 3DGS leads to difficulties in compression. To formulate unstructured attributes into normative distribution, we propose a well-structured tri-plane to encode Gaussian attributes, leveraging the distribution of attributes for compression. To exploit the correlations among adjacent Gaussians, K-Nearest Neighbors (KNN) is used when decoding Gaussian distribution from the Tri-plane. We also introduce Gaussian position information as a prior of the position-sensitive decoder. Additionally, we incorporate an adaptive wavelet loss, aiming to focus on the high-frequency details as iterations increase. Our approach has achieved results that are comparable to or surpass that of SOTA 3D Gaussians Splatting compression work in extensive experiments across multiple datasets. The codes are released at https://github.com/timwang2001/TC-GS.