pyVSR

Resources.PyVSR History

Hide minor edits - Show changes to markup

June 30, 2017 by 134.226.84.102 -
Deleted line 7:

(code available this weekend, in case you accessed the page early)

June 30, 2017 by 109.255.58.178 -
Changed line 36 from:
  * Implementation based on [Menpo](http://www.menpo.org/)
to:
  * Implementation based on Menpo
June 30, 2017 by 109.255.58.178 -
Changed lines 19-23 from:
  * currently supported: 
    * TCD-TIMIT
      * speaker-dependent protocol
      * speaker-independent protocol
      * single person
to:
  • currently supported:
    • TCD-TIMIT
      • speaker-dependent protocol
      • speaker-independent protocol
      • single person
Changed lines 25-30 from:
  * Discrete Cosine Transform (DCT)
    * Automatic ROI extraction
    * Configurable window size
    * Fourth order accurate derivatives
    * Sample rate interpolation
    * Storage in HDF5 format
to:
  • Discrete Cosine Transform (DCT)
    • Automatic ROI extraction
    • Configurable window size
    • Fourth order accurate derivatives
    • Sample rate interpolation
    • Storage in HDF5 format
Changed lines 32-36 from:
  * Active Appearance Models (AAM)
    * Do NOT require manually annotated landmarks
    * Face, lips, and chin models supported
    * Parameters obtainable either through fitting or projection
    * Implementation based on [Menpo](http://www.menpo.org/)
to:
  • Active Appearance Models (AAM)
    • Do NOT require manually annotated landmarks
    • Face, lips, and chin models supported
    • Parameters obtainable either through fitting or projection
    • Implementation based on [Menpo](http://www.menpo.org/)
Changed lines 38-39 from:
  * Point cloud of facial landmarks 
    * OpenFace wrapper
to:
  • Point cloud of facial landmarks
    • OpenFace wrapper
Changed lines 42-44 from:
  * easy HTK wrapper for Python 
  * optional bigram language model
  * multi-threaded support (both for training and decoding at full CPU Power)
to:
  • easy HTK wrapper for Python
  • optional bigram language model
  • multi-threaded support (both for training and decoding at full CPU Power)
Changed line 47 from:
  * pyVSR has a simple, modular, object-oriented architecture
to:
  • pyVSR has a simple, modular, object-oriented architecture
June 29, 2017 by 134.226.84.102 -
Added line 8:

(code available this weekend, in case you accessed the page early)

June 29, 2017 by 134.226.84.102 -
Changed lines 3-9 from:

pyVSR

Python toolkit for Visual Speech Recognition

Add markdown file

Link to github

to:

pyVSR

Python toolkit for Visual Speech Recognition

GitHub page

About

pyVSR is a Python toolkit aimed at running Visual Speech Recognition (VSR) experiments in a traditional framework (e.g. handcrafted visual features, Hidden Markov Models for pattern recognition).

The main goal of pyVSR is to easily reproduce VSR experiments in order to have a baseline result on most publicly available audio-visual datasets.

What can you do with pyVSR:

1. Fetch a filtered list of files from a dataset

  * currently supported: 
    * TCD-TIMIT
      * speaker-dependent protocol
      * speaker-independent protocol
      * single person

2. Extract visual features:

  * Discrete Cosine Transform (DCT)
    * Automatic ROI extraction
    * Configurable window size
    * Fourth order accurate derivatives
    * Sample rate interpolation
    * Storage in HDF5 format

  * Active Appearance Models (AAM)
    * Do NOT require manually annotated landmarks
    * Face, lips, and chin models supported
    * Parameters obtainable either through fitting or projection
    * Implementation based on [Menpo](http://www.menpo.org/)

  * Point cloud of facial landmarks 
    * OpenFace wrapper

3. Train Hidden Markov Models (HMMs)

  * easy HTK wrapper for Python 
  * optional bigram language model
  * multi-threaded support (both for training and decoding at full CPU Power)

4. Extend the support for additional features

  * pyVSR has a simple, modular, object-oriented architecture
June 29, 2017 by 134.226.84.102 -
Changed lines 3-4 from:

pyVSR

Python toolkit for Visual Speech Recognition

to:

pyVSR

Python toolkit for Visual Speech Recognition

June 29, 2017 by 134.226.84.102 -
Added lines 2-4:

pyVSR

Python toolkit for Visual Speech Recognition

June 29, 2017 by 134.226.84.102 -
Added lines 1-2:
Added line 4:
June 29, 2017 by 134.226.84.102 -
Changed line 1 from:

TODO

to:

Add markdown file

June 29, 2017 by 134.226.84.102 -
June 29, 2017 by 134.226.84.102 -
Added lines 1-2:

TODO Link to github

Page last modified on June 30, 2017