Skip to content

Course project for computer vision, implemented fine-grain image classification with CNN + SIFT + SVM.

Notifications You must be signed in to change notification settings

lizihan021/Dog-Breed-Classify

Repository files navigation

EECS 442: Fine-grained Dog Breed Classification

Lina Zhang, Wen He and Zihan Li

This project is intended as final project for EECS 442 Computer Vision. Our report is in the project folder.

Dependency: Keras, TensorFlow, OpenCV3, Python3

To run the program, download Columbia Dogs from here: ftp://ftp.umiacs.umd.edu/pub/kanazawa/CU_Dogs.zip. Put it in root folder and run python3 main.py.

Training procedure takes 1 hour on Nvidia GTX 1080 GPU.

About

Course project for computer vision, implemented fine-grain image classification with CNN + SIFT + SVM.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages