当前位置:网站首页>[2021]GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fields
[2021]GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fields
2022-07-05 06:17:00 【Dark blue blue blue】
NeRF in , We use a radiation field to simulate a real scene , Although the camera can change its position at will , But the things in the scene are fixed , Because a radiation field is a whole . In this paper, multiple radiation fields are combined , Thus, any radiation field can move freely in the scene . Of course , In this paper, the radiation field radiance field be called feature field.
The composition of the article :
1. First of all, the whole process is used GRAF That set , So it includes a generator and a discriminator , But there are differences inside the generator . One side GIRAFFE There are multiple radiation fields in , On the other hand, a little change has been made in rendering .
2. First , Because there are multiple radiation fields , Therefore, each radiation field has its own set of Networks , Corresponding to different positions , Shape and texture parameters .
3. The second is how to combine multiple radiation fields , It's easy , Is along the light , Sum the colors according to the density of each object .
4. Finally, how to render ,GRAF It's done directly with volume rendering , But this paper adds a neural rendering after volume rendering , It is estimated that it is to improve the image quality , It is equivalent to refining the picture . It is worth mentioning that , The result of volume rendering in this paper is not a graph in the usual sense , It's a feature map , Because it does not output color , It is a characteristic value . Personally think that , The reason for this is that there is also a neural renderer behind , Therefore, some of the expression ability is reserved .
5. The last discriminator and GRAF The same as in China , It is a network that simply judges whether the output image is true or synthetic .
边栏推荐
- 1039 Course List for Student
- 【LeetCode】Day95-有效的数独&矩阵置零
- [rust notes] 14 set (Part 1)
- Leetcode-6109: number of people who know secrets
- LeetCode 0108. Convert an ordered array into a binary search tree - the median of the array is the root, and the left and right of the median are the left and right subtrees respectively
- Appium automation test foundation - Summary of appium test environment construction
- Appium基础 — 使用Appium的第一个Demo
- 927. 三等分 模拟
- Leetcode-22: bracket generation
- Chapter 6 relational database theory
猜你喜欢
MySQL advanced part 1: index
阿里新成员「瓴羊」正式亮相,由阿里副总裁朋新宇带队,集结多个核心部门技术团队
2021apmcm post game Summary - edge detection
MySQL advanced part 2: MySQL architecture
Appium automation test foundation - Summary of appium test environment construction
LeetCode 0108. Convert an ordered array into a binary search tree - the median of the array is the root, and the left and right of the median are the left and right subtrees respectively
1.13 - RISC/CISC
阿里巴巴成立企业数智服务公司“瓴羊”,聚焦企业数字化增长
LVS简介【暂未完成(半成品)】
快速使用Amazon MemoryDB并构建你专属的Redis内存数据库
随机推荐
JS quickly converts JSON data into URL parameters
leetcode-1200:最小绝对差
【Rust 笔记】13-迭代器(下)
C - XOR to all (binary topic)
Usage scenarios of golang context
A reason that is easy to be ignored when the printer is offline
做 SQL 性能优化真是让人干瞪眼
阿里新成员「瓴羊」正式亮相,由阿里副总裁朋新宇带队,集结多个核心部门技术团队
Règlement sur la sécurité des réseaux dans les écoles professionnelles secondaires du concours de compétences des écoles professionnelles de la province de Guizhou en 2022
QQ电脑版取消转义符输入表情
Matrixdb V4.5.0 was launched with a new mars2 storage engine!
Leetcode recursion
[rust notes] 16 input and output (Part 1)
LeetCode-61
MySQL advanced part 1: stored procedures and functions
leetcode-31:下一个排列
liunx启动redis
Introduction to LVS [unfinished (semi-finished products)]
leetcode-9:回文数
1.13 - RISC/CISC