3D Gaussian Splatting

作为图形学当下最热的方向之一（神经网络结合），感觉始终无法绕开这个坎，遂决定仔细研读一下这篇史诗巨作。抛雪球算法连传统的mesh都摒弃了，算是开辟了一个非常热的赛道。

题目：real time 的含金量！Radiance Field Rendering，这个应该已经有铺垫了？

摘要：

1. Radiance Field methods have recently revolutionized novel-view synthesis of scenes captured with multiple photos or videos.

革命性的新方法！开篇直接开始介绍背景。

2. However, achieving high visual quality still requires neural networks that are costly to train and render, while recent faster methods inevitably trade off speed for quality.

然而，经典的进行转折。该方法没办法兼顾快和高质量。并没有全盘否定别人，只是侧面暗示自己的优点。

3. For unbounded and complete scenes (rather than isolated objects) and 1080p resolution rendering, no current method can achieve real-time display rates. 

作为对上一句话的介绍，谈谈该方法在人眼能分辨的1080p上的现实情况。（举例论证）

4. We introduce three key elements that allow us to achieve state-of-the-art visual quality while maintaining competitive training times and importantly allow high-quality real-time (≥ 30 fps) novel-view synthesis at 1080p resolution.

创新点有3，哥们在这边全给你列出来。自然而然的后半句讲实验的总述性结果，给人一个小小的震撼。（这个真嘟，非常之有用。别人审稿人看论文摘要肯定是重点，你摘要没有亮点或很nb的结论或者新颖方法（其实也是要服务于结论的），根本没人会理。接下来就是分三点介绍自己的工作啦~

5. First, starting from sparse points produced during camera calibration, we represent the scene with 3D Gaussians that preserve desirable properties of continuous volumetric radiance fields for scene optimization while avoiding unnecessary computation in empty space;

首先，基于稀疏点，我们用3D高斯保存场景。（核心之核心，场景的保存格式从mesh变成椭球）。当然，这种表示很有效，可以避免冗余的计算。

6. Second, we perform interleaved optimization/density control of the 3D Gaussians, notably optimizing anisotropic covariance to achieve an accurate representation of the scene;

优化各向异性协方差，提高精度。

7. Second, we perform interleaved optimization/density control of the 3D Gaussians, notably optimizing anisotropic covariance to achieve an accurate representation of the scene;

开发了一个抛雪球算法，用于实时渲染。核心之一，关于如何将高斯椭球转变为屏幕上的像素。

作为场景重建以及一种新的渲染流程，真的很厉害。开篇的几幅图也明确的说明了效果之好。感叹大佬的代码/idea之牛。