• 设为首页
  • 点击收藏
  • 手机版
    手机扫一扫访问
    迪恩网络手机版
  • 关注官方公众号
    微信扫一扫关注
    迪恩网络公众号

dineshreddy91/Occlusion_Net: [CVPR2019]Occlusion-Net: 2D/3D Occluded Keypoint Lo ...

原作者: [db:作者] 来自: 网络 收藏 邀请

开源软件名称(OpenSource Name):

dineshreddy91/Occlusion_Net

开源软件地址(OpenSource Url):

https://github.com/dineshreddy91/Occlusion_Net

开源编程语言(OpenSource Language):

Python 98.5%

开源软件介绍(OpenSource Introduction):

Occlusion-Net: 2D/3D Occluded Keypoint Localization Using Graph Networks

N Dinesh Reddy, Minh Vo, Srinivasa G. Narasimhan

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.

[Project] [Paper] [Supp] [Bibtex ]

More Results


Result of Occlusion-Net on a live video from youtube

Installation

Setting up with docker

All the stable releases of docker-ce installed from https://docs.docker.com/install/

Install the nvidia-docker from https://github.com/NVIDIA/nvidia-docker

Setting up the docker

nvidia-docker build -t occlusion_net .

Setting up data

You need to fill the Access Form to get a email regarding the dataset and setup at using the following commands:

git clone https://github.com/dineshreddy91/carfusion_to_coco
cd carfusion_to_coco
virtualenv carfusion2coco -p python3.6
source carfusion2coco/bin/activate
pip install cython numpy
pip install -r requirements.txt
python download_carfusion.py (This file need to be downloaded by requesting, please fill to get access to the data)
sh carfusion_coco_setup.sh
deactivate

The final folder format to train on carfusion data needs to look :

Occlusion-Net
   └─datasets
       └─carfusion
           └─train
               └─car_craig1
                   └───images
                       01_0000.jpg
                       01_0001.jpg
                       ...   
                   └───bb
                      01_0000.txt
                      01_0001.txt
                      ...
                   └───gt
                      01_0000.txt   
                      01_0001.txt
                      ...
           └─test
               └─car_penn1
                   └───images
                       01_0000.jpg
                       01_0001.jpg
                       ...   
                   └───bb
                      01_0000.txt
                      01_0001.txt
                      ...
                   └───gt
                      01_0000.txt   
                      01_0001.txt
                      ...
           └─annotations
               car_keypoints_train.json
               car_keypoints_test.json
               

Running with docker

Training the model on the carfusion dataset

sh train.sh occlusion_net <Path_to_Carfusion_dataset>

Testing on a sample image

Download a pretrained model from [Google Drive]

Results on a sample demo image

sh test.sh occlusion_net demo/demo.jpg

Citation

@inproceedings{onet_cvpr19,
author = {Reddy, N. Dinesh and Vo, Minh and Narasimhan, Srinivasa G.},
title = {Occlusion-Net: 2D/3D Occluded Keypoint Localization Using Graph Networks},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
pages = {7326--7335},
year = {2019}
}
@InProceedings{Reddy_2018_CVPR,
author = {Dinesh Reddy, N. and Vo, Minh and Narasimhan, Srinivasa G.},
title = {CarFusion: Combining Point Tracking and Part Detection for Dynamic 3D Reconstruction of Vehicles},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2018}
}



鲜花

握手

雷人

路过

鸡蛋
该文章已有0人参与评论

请发表评论

全部评论

专题导读
上一篇:
Quantower/Localization发布时间:2022-08-15
下一篇:
mihaidusmanu/cross-descriptor-vis-loc-map: Cross-Descriptor Visual Localization ...发布时间:2022-08-15
热门推荐
阅读排行榜

扫描微信二维码

查看手机版网站

随时了解更新最新资讯

139-2527-9053

在线客服(服务时间 9:00~18:00)

在线QQ客服
地址:深圳市南山区西丽大学城创智工业园
电邮:jeky_zhao#qq.com
移动电话:139-2527-9053

Powered by 互联科技 X3.4© 2001-2213 极客世界.|Sitemap