Skip to content

fashion classification project with Mask R-CNN (work in progress)

License

Notifications You must be signed in to change notification settings

yyzz1010/Image_Recognition_ClosetAI

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Image Recognition: ClosetAI

Combining the techniques of Object Detection and Instance Segmentation, using deep learning model Mask R-CNN in clothes retrieval. Various medias such as clothing image and street fashion video can be analyzed by the model. Also, ClosetAI can generate business intelligence analysis in giving you a popular fashion items report.

Data Source

DeepFashion2 Dataset which contains including 191,961 training images and 32,153 validation images. Json files of image information such as annotations and category names are also provided.

Exploratory Data Analysis

Data Preprocessing

Deployment

Skills Acquired

  • Deep Learning:
  • Git LFS: upload large file greater than 100MB to GitHub using terminal command, self-written tutorial can be found here

Citations

@article{DeepFashion2,
  author = {Yuying Ge and Ruimao Zhang and Lingyun Wu and Xiaogang Wang and Xiaoou Tang and Ping Luo},
  title={A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images},
  journal={CVPR},
  year={2019}
}
@misc{matterport_maskrcnn_2017,
  title={Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow},
  author={Waleed Abdulla},
  year={2017},
  publisher={Github},
  journal={GitHub repository},
  howpublished={\url{https://github.com/matterport/Mask_RCNN}},
}

Acknowledgements

This project cannnot be accomplished without references from below repositories and blog post. Thank you coders for sharing your experience! =]

About

fashion classification project with Mask R-CNN (work in progress)

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published