Skip to content

Latest commit

 

History

History
18 lines (13 loc) · 559 Bytes

README.md

File metadata and controls

18 lines (13 loc) · 559 Bytes

deepfood

Classify images of food items using Deep Learning (CNN)

Simple Non Deep learning based approaches

  • Color histograms: code: histogram.py
  • Bag of Words: code: siftsvm.py

Deep learning based methods

  • Custom cnn model: custom.py
  • Vgg16 pretraining: bottleneck.py

How to Run?

Code is in python. So run using python filename.py Note that we use keras library for our CNN training and sklearn for non deep learning methods

Dataset

We collected food data by crawling: https://drive.google.com/drive/folders/0Bzvk20lZHSdYOWgwQktDX1lpcWs