• 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
    公众号

Tangshitao/Dense-Scene-Matching: Learning Camera Localization via Dense Scene Ma ...

原作者: [db:作者] 来自: 网络 收藏 邀请

开源软件名称(OpenSource Name):

Tangshitao/Dense-Scene-Matching

开源软件地址(OpenSource Url):

https://github.com/Tangshitao/Dense-Scene-Matching

开源编程语言(OpenSource Language):

Python 55.9%

开源软件介绍(OpenSource Introduction):

This repository contains code of our CVPR 2021 paper - "Learning Camera Localization via Dense Scene Matching" by Shitao Tang, Chengzhou Tang, Rui Huang, Siyu Zhu and Ping Tan.

This paper presents a new method for scene agnostic camera localization using dense scene matching (DSM), where a cost volume is constructed between a query image and a scene. The cost volume and the corresponding coordinates are processed by a CNN to predict dense coordinates. Camera poses can then be solved by PnP algorithms.

If you find this project useful, please cite:

@inproceedings{Tang2021Learning,
  title={Learning Camera Localization via Dense Scene Matching},
  author={Shitao Tang, Chengzhou Tang, Rui Huang, Siyu Zhu and Ping Tan},
  booktitle={Computer Vision and Pattern Recognition (CVPR)},
  year={2021}
}

Usage

Environment

  • The codes are tested along with
    • pytorch=1.4.0
    • lmdb (optional)
    • yaml
    • skimage
    • opencv
    • numpy=1.17
    • tensorboard

Installation

  • Build PyTorch operations
      cd libs/model/ops
      python setup.py install
    
  • Build PnP algorithm
      cd libs/utils/lm_pnp
      mkdir build
      cd build
      cmake ..
      make all
    

Train and Test

  • Download

    You can download the trained models and label files for 7scenes, Cambridge, Scannet.

    For 7scenes, you can use the prepared data in the following.

    Chess Fire Heads Office Pumpkin Kitchen Stairs

    For Cambridge landmarks, you can download image files here, and depths here.

  • Test

    Please refer to configs/7scenes.yaml for detailed explaination of how to set label file path and image file path.

    • 7scenes
      python tools/video_test.py --config configs/7scenes.yaml
      
    • Camrbrige
      python tools/video_test.py --config configs/cambridge.yaml
      
  • Train

    We use ResNet-FPN pretrained model.

      python tools/train_net.py
    



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
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