Giter Site home page Giter Site logo

chestnut-q / jittor-sequence-nvs Goto Github PK

View Code? Open in Web Editor NEW
1.0 1.0 0.0 2.66 MB

第二届计图人工智能挑战赛-赛题二:可微渲染新视角生成的提交代码

Shell 0.16% Python 3.33% C++ 77.11% C 3.71% CMake 2.12% HTML 0.17% Cuda 1.60% Fortran 11.62% XSLT 0.06% JavaScript 0.07% CSS 0.05%

jittor-sequence-nvs's Introduction

Jittor 可微渲染新视角生成 NeRF/NGP

简介

本项目为第二届计图人工智能挑战赛-赛题二:可微渲染新视角生成的开源代码。

队伍名称:sequence

队伍学校:清华大学

队伍人员:秦若愚(队长)、李正远

环境配置

本项目的所有训练及生成过程均可在 1 张 3090 GPU 上完成。由于针对不同场景使用了不同的算法,因此运行环境共有 2 套,下面分别介绍。

NeRF

运行环境

  • ubuntu 20.04 LTS
  • python >= 3.8
  • jittor >= 1.3.3

安装依赖

在 NeRF 目录下执行以下命令

python -m pip install -r requirements.txt

NGP

运行环境

  • ubuntu 20.04 LTS
  • python >= 3.8
  • jittor >= 1.3.3

安装依赖

在 NGP 目录下执行以下命令

sudo apt-get install tcl-dev tk-dev python3-tk
python -m pip install -r requirements.txt
cd python
python -m pip install -e .

训练&生成

Easyship

首先将 NeRF/configs/Easyship.txt 中 basedir(输出目录)、datadir(数据目录)修改为合适的目录,然后在 NeRF 环境下,进入 NeRF 目录并运行以下命令

python nerf.py --config ./configs/Easyship.txt

运行完成后便可在 basedir 下看到训练的 checkpoints 和生成结果。

Car

首先将 NeRF/configs/Car.txt 中 basedir、datadir 修改为合适的目录,然后在 NeRF 环境下,进入 NeRF 目录并运行以下命令

python nerf.py --config ./configs/Car.txt

运行完成后便可在 basedir 下看到训练的 checkpoints 和生成结果。

Coffee

首先将 NGP/projects/ngp/configs/coffee.py 中 dataset_dir(数据目录)、log_dir(输出目录)修改为合适的目录,然后在 NGP 环境下,进入 NGP 目录并运行以下命令

bash bash/run_coffee.sh

运行完成后便可在 log_dir下看到生成结果。

Scar

首先将 NGP/projects/ngp/configs/scar.py 中 dataset_dir、log_dir 修改为合适的目录,然后在 NGP 环境下,进入 NGP 目录并运行以下命令

bash bash/run_scar.sh

运行完成后便可在 log_dir下看到生成结果。

Scarf

首先将 NGP/projects/ngp/configs/scarf.py 中 dataset_dir、log_dir 修改为合适的目录,然后在 NGP 环境下,进入 NGP 目录并运行以下命令

bash bash/run_scarf.sh

运行完成后便可在 log_dir下看到生成结果。

致谢

代码框架

本项目代码基于以下两个项目完成:

  1. https://github.com/Jittor/jrender
  2. https://github.com/Jittor/JNeRF

参考文献

  1. Mildenhall B, Srinivasan P P, Tancik M, et al. Nerf: Representing scenes as neural radiance fields for view synthesis[C]//European conference on computer vision. Springer, Cham, 2020: 405-421.
  2. MüllerT, Evans A, Schied C, et al., Instant Neural Graphics Primitives with a Multiresolution Hash Encoding, arXiv: 2201.05989, 2022.

jittor-sequence-nvs's People

Contributors

chestnut-q avatar

Stargazers

liuchu avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.