arXiv:2511.13264v2 Announce Type: replace 
Abstract: 3D Gaussian Splatting has emerged as a transformative technique in novel view synthesis, primarily due to its high rendering speed and photorealistic fidelity. However, its memory footprint scales rapidly with scene complexity, often reaching several gigabytes. Existing methods address this issue by introducing compression strategies that exploit primitive-level redundancy through similarity detection and quantization. We aim to surpass the compression limits of such methods by incorporating symmetry-aware techniques, specifically targeting mirror symmetries to eliminate redundant primitives. We propose a novel compression framework, SymGS, introducing learnable mirrors into the scene, thereby eliminating local and global reflective redundancies for compression. Our framework functions as a plug-and-play enhancement to state-of-the-art compression methods, (e.g. HAC) to achieve further compression. Compared to HAC, we achieve $1.66 \times$ compression across benchmark datasets (upto $3\times$ on large-scale scenes). On an average, SymGS enables $\bf{108\times}$ compression of a 3DGS scene, while preserving rendering quality. The project page and supplementary can be found at symgs.github.io

تقدم الورقة SymGS، وهو إطار جديد لضغط بيانات 3D Gaussian Splatting من خلال الاستفادة من التماثلات المحلية. تهدف هذه الطريقة إلى تقليل حجم الذاكرة المرتبطة بالمشاهد المعقدة، والتي يمكن أن تصل إلى عدة جيجابايت. من خلال إدخال مرايا قابلة للتعلم لإزالة العناصر الزائدة، يحسن SymGS طرق الضغط الحالية، محققًا تحسينات كبيرة في الكفاءة مقارنة بالتقنيات التقليدية.

El artículo presenta SymGS, un nuevo marco para comprimir datos de 3D Gaussian Splatting aprovechando simetrías locales. Este enfoque busca reducir la huella de memoria asociada con escenas complejas, que pueden alcanzar varios gigabytes. Al introducir espejos aprendibles para eliminar primitivas redundantes, SymGS mejora los métodos de compresión existentes, logrando mejoras significativas en eficiencia en comparación con técnicas tradicionales.

L'article présente SymGS, un nouveau cadre pour compresser les données de 3D Gaussian Splatting en exploitant les symétries locales. Cette approche vise à réduire l'empreinte mémoire associée aux scènes complexes, qui peuvent atteindre plusieurs gigaoctets. En introduisant des miroirs apprenables pour éliminer les primitives redondantes, SymGS améliore les méthodes de compression existantes, réalisant des gains significatifs en efficacité par rapport aux techniques traditionnelles.

The paper presents SymGS, a novel framework for compressing 3D Gaussian Splatting data by leveraging local symmetries. This approach aims to reduce the memory footprint associated with complex scenes, which can reach several gigabytes. By introducing learnable mirrors to eliminate redundant primitives, SymGS enhances existing compression methods, achieving significant improvements in efficiency compared to traditional techniques.

SymGS : Leveraging Local Symmetries for 3D Gaussian Splatting Compression

Was this article worth reading? Share it

SwapAnything.io

SVGenius

Sprello

pikaai.org

SVGX

IMGFX.DEV

Ready to build your own newsroom?