Image-GS: Content-Adaptive Image Representation via 2D Gaussians
[**Yunxiang Zhang**](https://yunxiangzhang.github.io/)
1\*,
[**Bingxuan Li**](https://bingxuan-li.github.io/)
1\*,
[**Alexandr Kuznetsov**](https://alexku.me/)
3†,
[**Akshay Jindal**](https://www.akshayjindal.com/)
2,
[**Stavros Diolatzis**](https://www.sdiolatz.info/)
2,
[**Kenneth Chen**](https://kenchen10.github.io/)
1,
[**Anton Sochenov**](https://www.intel.com/content/www/us/en/developer/articles/community/gpu-researchers-anton-sochenov.html)
2,
[**Anton Kaplanyan**](http://kaplanyan.com/)
2,
[**Qi Sun**](https://qisun.me/)
1
\* Equal contribution † Work done while at Intel
1
2
3
Neural image representations have emerged as a promising approach for encoding and rendering visual data. Combined with learning-based workflows, they demonstrate impressive trade-offs between visual fidelity and memory footprint. Existing methods in this domain, however, often rely on fixed data structures that suboptimally allocate memory or compute-intensive implicit models, hindering their practicality for real-time graphics applications.
Inspired by recent advancements in radiance field rendering, we introduce Image-GS, a content-adaptive image representation based on 2D Gaussians. Leveraging a custom differentiable renderer, Image-GS reconstructs images by adaptively allocating and progressively optimizing a group of anisotropic, colored 2D Gaussians. It achieves a favorable balance between visual fidelity and memory efficiency across a variety of stylized images frequently seen in graphics workflows, especially for those showing non-uniformly distributed features and in low-bitrate regimes. Moreover, it supports hardware-friendly rapid random access for real-time usage, requiring only 0.3K MACs to decode a pixel. Through error-guided progressive optimization, Image-GS naturally constructs a smooth level-of-detail hierarchy. We demonstrate its versatility with several applications, including texture compression, semantics-aware compression, and joint image compression and restoration.
Figure 1: Image-GS reconstructs an image by adaptively allocating and progressively optimizing a set of colored 2D Gaussians. It achieves favorable rate-distortion trade-offs, hardware-friendly random access, and flexible quality control through a smooth level-of-detail stack. (a) visualizes the optimized spatial distribution of Gaussians (20% randomly sampled for clarity). (b) Image-GS’s explicit content-adaptive design effectively captures non-uniformly distributed image features and better preserves fine details under constrained memory budgets. In the inset error maps, brighter colors indicate larger errors.