Related Works

这一部分是由很多小节构成，小节之间可以没有连贯性。在这里举例一个小节的多种写作方式，可以自由组合调整

先笼统介绍，再细说一些方法的核心区别
Some using full-supervision in the form of semantic labels, others find meaningful directions in a self-supervised fashion, and, finally, recent works present unsupervised methods to achieve the same goal.
More specifically, xx use supervision in the form of facial attribute labels to find meaningful linear direction... yy perform eigenvector decomposition on the generator's weights to find edit directions without additional supervision.

一句话大概说说这个领域有很多方法
The most related to our work can be divided into two categories: A and B. The first focuses on xx ; The second xx

针对每一个方法，先概述数量多少
Only very few works have explored xx in the context of xx

有很多方法为了解决什么问题，但他们都只关心了。。
We review recent advances in xx, as well as xx.
xx are related to many other bodies of work.
Our method build upon this line of research
Its formulation has been adapted for improving A, dealing with B, removing C, and handling D. Although there have been some early attempts in E, the usage of F has never been explored for G. This work addresses this gap by proposing H as I that leverages K.
Towards this goal

同期工作

最后做个比较
The concurrently developed 3D-aware GANs StyleNeRF and CIPS-3D [73] demonstrate impressive image quality. The central distinction between these and ours is that while StyleNeRF and CIPS-3D operate primarily in image-space, with less emphasis on the 3D representation, our method operates primarily in 3D. Our approach demonstrates greater view consistency, and is capable of generating high-quality 3D shapes.