Mip-Splatting: Alias-Free 3D Gaussian Splatting
By Zehao Yu, Anpei Chen, Binbin Huang, Torsten Sattler, Andreas Geiger. Computer Vision and Pattern Recognition, 2023
860 citations recorded by the source corpus.
Recently, 3D Gaussian Splatting has demonstrated impressive novel view synthesis results, reaching high fidelity and efficiency. However, strong artifacts can be observed when changing the sampling rate, e.g., by changing focal length or camera distance. We find that the source for this phenomenon can be attributed to the lack of 3D frequency constraints and the usage of a 2D dilation filter. To address this problem, we introduce a 3D smoothing filter to constrains the size of the 3D Gaussian primitives based on the maximal sampling frequency induced by the input views. It eliminates high-frequency artifacts when zooming in. Moreover, replacing 2D dilation with a 2D Mip filter, which simulates a 2D box filter, effectively mitigates aliasing and dilation issues. Our evaluation, including scenarios such a training on single-scale images and testing on multiple scales, validates the effectiveness of our approach.
Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design
View Paper: https://www.semanticscholar.org/paper/7e305e37b49feb8899ae94f641c4f2d0d01ef054
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computational Mechanics
In Computer Graphics and Computer-Aided Design
Citations in the corpus
- Octree-GS: Towards Consistent Real-time Rendering with LOD-Structured 3D Gaussians (2024, methodology): “3) Baselines: We compare our method against 2D-GS [15], 3D-GS [5], Scaffold-GS [3], Mip-Splatting [14] and two concurrent works, CityGaussian [19] and Hierarchical-GS [2].”
- gsplat: An Open-Source Library for Gaussian Splatting (2024, positive): “Mip-Splatting (Yu et al., 2024) proposes a low pass filter on projected 2D Gaussian covariances, ensuring a Gaussian’s extent always spans a pixel. gsplat supports rendering with the 2D anti-aliasing mode introduced in Yu et al.. Definitions are found in A.4”
- Structured 3D Latents for Scalable and Versatile 3D Generation (2024, neutral): “Figure XII and XIII provide two supplementary visualizations of complex scenes constructed with assets from our model, demonstrating its potential for production use.”
- Recent advances in 3D Gaussian splatting (2024, neutral): “Splatting [31] observed that changing the sampling rate, for example, the focal length, can greatly influence the quality of rendered images by introducing high-frequency Gaussian shape-like artifacts or strong dilation effects.”
- A Survey on 3D Gaussian Splatting (2024, neutral): “Second, researchers have developed methods for compressing Gaussian’s properties [31, 124, 183].”
- Continuous 3D Perception Model with Persistent State (2025, neutral): “…replacing hand-crafted components with learning-based alternatives (e.g., substituting conventional feature descriptors [6, 47] with learned ones [17, 20, 63, 73]), integrating data-driven priors [13, 77–79, 93, 97, 105] into the systems, or optimizing entire systems end-to-end [76, 78, 83, 94].”