Classifying model for Sign Language Recognition System
Project description
gesture-classifying-model
----------------
To use, first create GestureClassifier object::
>>> predictor = GestureClassifier()
This may take a while as model is being downloaded and loaded (approx. 86 MB of weights).
After creation, you can use the predictor, by passing tensor of frames to transform function.
Numpy array of passed images should have same number of channels, same width and height.
GestureClassifier receives tensor in following shape (None, 64, 64, 3). By that I mean::
>>> input_tensor.shape
>>> (any_num, 64, 64, 3),
where 'any_num' is any number of frames passed. Images should be in range of values 0-255.
You can finally predict probabilities of frame of cropped hands (letters are indexed in order shown below)::
>>> predictor.transform(input_tensor_of images)
Letters order::
'abcdefghiklmnopqrstuvwxy' # alphabetical order for indices from 1 to 24 without
----------------
To use, first create GestureClassifier object::
>>> predictor = GestureClassifier()
This may take a while as model is being downloaded and loaded (approx. 86 MB of weights).
After creation, you can use the predictor, by passing tensor of frames to transform function.
Numpy array of passed images should have same number of channels, same width and height.
GestureClassifier receives tensor in following shape (None, 64, 64, 3). By that I mean::
>>> input_tensor.shape
>>> (any_num, 64, 64, 3),
where 'any_num' is any number of frames passed. Images should be in range of values 0-255.
You can finally predict probabilities of frame of cropped hands (letters are indexed in order shown below)::
>>> predictor.transform(input_tensor_of images)
Letters order::
'abcdefghiklmnopqrstuvwxy' # alphabetical order for indices from 1 to 24 without
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Close
Hashes for gesture_classifying_model-0.1.9.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 6c4ee6550943d855ba62e4e6ee2841ed29b48d7397ecbf0a4bb81464423a9ac3 |
|
MD5 | e6ca61fffea7fe182a625553184fb1d0 |
|
BLAKE2b-256 | a5f7004d40275471d6e1da7c0baaa9ebaa29ef12c53a087d2843ef3fe67f9c79 |