# Structure-Guided Allocation of 2D Gaussians for Image Representation and Compression

arXiv: https://arxiv.org/abs/2512.24018 · HTML: https://mlsanity.com/paper/2512.24018/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Dec 2025.

## Built on by (most cited first, 5 of 5)

- [Adaptive Learned Image Compression with Graph Neural Networks](https://mlsanity.com/paper/2603.25316/) (2026)
- [AIR: Amortized Image Reconstruction Framework for Self-Supervised Feed-Forward 2D Gaussian Splatting](https://mlsanity.com/paper/2605.20820/) (2026)
- [Locality-Aware Density Control for Efficient Gaussian-based Image Representation](https://mlsanity.com/paper/2607.17896/) (2026)
- [Rate-Distortion Adaptive Primitive Selection for Omnidirectional Gaussian Splatting](https://mlsanity.com/paper/2609.34367/) (2026)
- [Recasting DiT for Extreme Image Compression with Hierarchical Visual Priors](https://mlsanity.com/paper/2610.18607/) (2026)

## Abstract (arXiv)

Recent advances in 2D Gaussian Splatting (2DGS) have demonstrated its potential as a compact image representation with millisecond-level decoding. However, existing 2DGS-based pipelines allocate representation capacity and parameter precision largely oblivious to image structure, limiting their rate-distortion (RD) efficiency at low bitrates. To address this, we propose a structure-guided allocation principle for 2DGS, which explicitly couples image structure with both representation capacity and quantization precision, while preserving native decoding speed. First, we introduce a structure-guided initialization that assigns 2D Gaussians according to spatial structural priors inherent in natural images, yielding a localized and semantically meaningful distribution. Second, during quantization-aware fine-tuning, we propose adaptive bitwidth quantization of covariance parameters, which grants higher precision to small-scale Gaussians in complex regions and lower precision elsewhere, enabling RD-aware optimization, thereby reducing redundancy without degrading edge quality. Third, we impose a geometry-consistent regularization that aligns Gaussian orientations with local gradient directions to better preserve structural details. Extensive experiments demonstrate that our approach substantially improves both the representational power and the RD performance of 2DGS while maintaining over 1000 FPS decoding. Compared with the baseline GSImage, we reduce BD-rate by 43.44% on Kodak and 29.91% on DIV2K.