Skip to content

This is a program for the EECS 504 final project in university of Michigan-Ann Arbor. In this program, we, CVDancer group build a pipeline to do the classification of smoking and calling based on deep learning.

Notifications You must be signed in to change notification settings

wangmiaowei/CVDancer_Classification

Repository files navigation

CVDancer_Classification

This a EECS 504 project built by CVdancer group in university of Michigan. Video(https://www.youtube.com/watch?v=hez54R_Hucs&feature=youtu.be) Paper report(https://arxiv.org/abs/2012.08026v1) The code is based on the two great work about image augmentation and classfication respectively.

Our objective is to create a system to classify the behaviors of smoking and calling. We achieved this using the InceptionV3 architecture.

The split_train_val.py and warmup.py are relics of early on in the project's life. As it stands, the code should not require this data to function as intended. The weight downloading link: https://files.slack.com/files-pri/T01GJ9C2L7M-F01GT0H53TM/download/weights2.h5

About

This is a program for the EECS 504 final project in university of Michigan-Ann Arbor. In this program, we, CVDancer group build a pipeline to do the classification of smoking and calling based on deep learning.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages