Skip to content

RitaMarques/Project-DeepLearning

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

97 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Project-DeepLearning - Identify American Sign Language in images


Hello in American Sign Language

In this project we try to identify the gestures in the American Sign Language Alphabet using a dataset of somewhat low resolution older pictures. Data is pulled directly in the notebook and preprocessing is also made available.

We further analyse which symbols are easier to identify and which are easier to confound and over this try different approaches to improve a base model.

We use a CNN based model, to identify the signals in images, with over 98% accuracy on the test images, but given the limited nature of training samples, new test samples with different conditions are not so easily identified.
The final report is available including other studies on the behaviour of the model.

Image dataset reference: Pugeault, N., and Bowden, R. (2011). Spelling It Out: Real-Time ASL Fingerspelling Recognition In Proceedings of the 1st IEEE Workshop on Consumer Depth Cameras for Computer Vision, jointly with ICCV'2011

About

Identify American Sign Language in images

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages