Summary of Poison-splat: Computation Cost Attack on 3d Gaussian Splatting, by Jiahao Lu et al.
Poison-splat: Computation Cost Attack on 3D Gaussian Splatting
by Jiahao Lu, Yifan Zhang, Qiuhong Shen, Xinchao Wang, Shuicheng Yan
First submitted to arxiv on: 10 Oct 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Cryptography and Security (cs.CR); Graphics (cs.GR); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper reveals a significant security vulnerability in 3D Gaussian splatting (3DGS), a dominant 3D representation technique. Researchers develop an attack named Poison-splat, which can maliciously tamper with input data to increase the computation memory and time needed for 3DGS training, potentially causing Denial-of-Service (DoS) attacks on servers. The attack is achieved by solving a bi-level optimization problem using tailored strategies. This vulnerability has been largely overlooked in 3DGS systems, highlighting the need for attention to this crucial security issue. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper shows that 3D Gaussian splatting (3DGS) has a hidden weakness. Someone can secretly change the pictures used to train the algorithm, making it take much longer and use more memory. This can even cause servers to stop working completely! The researchers created an attack called Poison-splat, which makes the algorithm slower and more memory-hungry. They hope that this discovery will make people pay attention to the security of 3DGS systems. |
Keywords
* Artificial intelligence * Attention * Optimization
