Giter Club home page Giter Club logo

3dcnn_hmm's Introduction

Purpose

This is the code the challenge"Chalearn Looking at People 2014โ€œ.


Gist:3DCNN(3D Convolutional Neural Networks) + Hidden Markov Networks


by Di WU: [email protected], 2014/05/27

Disclaimer

Since I have little time to improve the performance, hence the project files are all over the place/messy. Apologies in advance if that causes any inconvenience for reading and please contact me for missing modules if there is any.


Dependency:


(1) Nvidia enabled GPU

(2) scikit learn: for Depth image template matching task: http://scikit-learn.org/stable/

(3) opencv for python: for depth image scaling and shifting: http://docs.opencv.org/trunk/doc/py_tutorials/py_tutorials.html

(4) scipy for imresize and some reading matrix functionality

(5) cuda-convnet: for 3D convolutional neural networks (https://code.google.com/p/cuda-convnet/) Note: compiling the library is kind of complicated for windows(and the library needs GPU on your computer) I have a blog about compiling the cuda library under windows environment if ever needed:(http://vision.group.shef.ac.uk/wordpress/?p=1)

Train

To train the network, you first need to extract the relevant information and store the intermediate result:

  1. 'Step1_3DCNN_template.py' --> extract template

  2. 'Step1_3DCNN_batch.py' and'Step1_3DCNN_neutral_batch.py' --> extract normalized resize images using template matching method (could also be the 'Step1_3DCNN_normalize_by_sk.py' and 'Step1_3DCNN_sk_normlize_neutral.py' using skeleton for normalization

  3. 'Step1_transition_matrix.py' --> extract Transitional Matrix and Prior for Hidden Markov Models

  4. 'Main_3DCNN.py' --> the main file for training the network.

Thank you!

3dcnn_hmm's People

Contributors

stevenwudi avatar

Watchers

 avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.