Coder Social home page Coder Social logo

monodetr_paddle's Introduction

使用文档

环境要求

  • PaddlePaddle 2.4及以上
  • OS 64位操作系统
  • Python 3(3.5.1+/3.6/3.7/3.8/3.9),64位版本
  • pip/pip3(9.0.1+),64位版本
  • CUDA >= 10.1
  • cuDNN >= 7.6

安装说明

1. 安装PaddlePaddle

# CUDA10.2
python -m pip install paddlepaddle-gpu==0.0.0.post102 -f https://www.paddlepaddle.org.cn/whl/linux/gpu/develop.html

请确保您的PaddlePaddle安装成功。使用以下命令进行验证。

# 在您的Python解释器中确认PaddlePaddle安装成功
>>> import paddle
>>> paddle.utils.run_check()

# 确认PaddlePaddle版本
python -c "import paddle; print(paddle.__version__)"

注意

  1. 如果您希望在多卡环境下使用PaddleDetection,请首先安装NCCL

2. 编译自定义算子

# 安装依赖
git clone https://github.com/PaddlePaddle/Paddle3D.git
cd Paddle3D
pip install -r requirements.txt
pip install -e .

cd paddle3d/ops/
python setup.py install

cd ../../../

2. 安装环境

# 安装依赖
git clone https://github.com/wangna11BD/MonoDETR_paddle.git 

cd MonoDETR_paddle
pip install -r requirements.txt
mkdir logs
  1. 下载 KITTI 数据集,数据集结构如下:
    │MonoDETR_paddle/
    ├──...
    ├──data/KITTIDataset/
    │   ├──ImageSets/
    │   ├──training/
    │   ├──testing/
    ├──...
    

使用说明

MonoDETR模型的配置文件为configs/monodetr.yaml目录下

resnet50预训练模型下载

下载预训练模型放在当前文件夹下 https://ecloud.baidu.com?t=e34a3edaf4cd7160ef09995bef241171

训练

bash train.sh configs/monodetr.yaml > logs/monodetr.log

评估

bash test.sh configs/monodetr.yaml

复现结果

训练环境:16G P40 cuda10.2 py3.6.8 torch1.12.0 paddle2.4 复现结果对比 (paddle1为第一版代码训练结果,paddle2为第二版代码训练结果)

Easy Mod. Hard log 模型
torch 25.46% 19.74% 16.57% log model
torch 25.77% 18.63% 15.38% - -
paddle1 26.47% 18.78% 15.54% - -
paddle1 26.21% 18.93% 15.75% - -
paddle1 27.23% 19.55% 16.15% log1log2 model
paddle1 23.91% 18.10% 15.13% - -
paddle2 26.29% 19.23% 16.05% log model
paddle2 26.04% 19.30% 15.92% log model
paddle2 25.93% 19.24% 16.09% log model
paddle2 26.57% 19.12% 15.90% log model
paddle2 26.33% 19.09% 15.89% log model

Citation

@article{zhang2022monodetr,
  title={MonoDETR: Depth-aware Transformer for Monocular 3D Object Detection},
  author={Zhang, Renrui and Qiu, Han and Wang, Tai and Xu, Xuanzhuo and Guo, Ziyu and Qiao, Yu and Gao, Peng and Li, Hongsheng},
  journal={arXiv preprint arXiv:2203.13310},
  year={2022}
}

monodetr_paddle's People

Contributors

wangna11bd avatar

Stargazers

 avatar  avatar  avatar  avatar

Watchers

 avatar

monodetr_paddle's Issues

update the model file link

Hello, thank you for your open source, I noticed that your model file link is invalid, can you update the link? thank you

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.