Image-classification-and-localization

Image classifier for the VOC-2010 Dataset with and without Transfer Learning. Deep Convolutional Networks were used for classification and localization.

Packages and Dependencies Required :-
Keras (Tensorflow Background)
matplotlib
PIL (pillow)
jupyter notebook
numpy
pickle

-> NNfunctions.py is a program with all the utility functions used for Dataset creation and Image localization.

-> Change the image folder paths to wherever your image dataset is stored .

For furthur details on Architecture, results and transfer learning implementation, refer "report.pdf" .

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
SSD		SSD
pics		pics
Custom_classifier_detector.ipynb		Custom_classifier_detector.ipynb
FINAL NOTEBOOK.ipynb		FINAL NOTEBOOK.ipynb
NNfunctions.py		NNfunctions.py
README.md		README.md
TransferLearning_boundingBox.ipynb		TransferLearning_boundingBox.ipynb
build_dataset.py		build_dataset.py
build_input.py		build_input.py
classification.py		classification.py
confusion_matrix.py		confusion_matrix.py
generator.py		generator.py
get_data_from_XML.py		get_data_from_XML.py
localization.py		localization.py
report.pdf		report.pdf
vgg_bounding_box.ipynb		vgg_bounding_box.ipynb
vgg_classifier.ipynb		vgg_classifier.ipynb

vishwassathish/Image-classification-and-localization

Folders and files

Latest commit

History

Repository files navigation

Image-classification-and-localization

About

Topics

Resources

Stars

Watchers

Forks

Languages