rshrc / gait-recognition

Distance Recognition of a Human Being with Deep CNN's

Cross-View Gait Based Human Identification

GAIT is a kind of behavioral biometric feature, whose raw data are video sequences presenting walking people. It is particularly suitable for long-distance human identification, and requires no explicit co-operation by subjects, compared with other kinds of biometric features such as fingerprint and iris.

Using multiple images of the same person from different angles.

One person having multiple outfits

The Deep Neural Network Architecture

Currently Under Active Development

gait-recognition's Issues

Dataset?

Which dataset are you guys using, public or private?

Im newbie here, so can you tell me how to run this project?

Using the Deformable Conv. Neural Networks to enhance and optimize the existing CNN

Using a to increase the accuracy and thereby enhance/optimize the existing implementation of the Keras based CNN used in this project.

model.h5?

YOLO for Live Web Cam

Currently yoloutils/detection.py can only work on static image files

It should also be able to work on Live Video Feed from the Web Cam

No alternatives considered

Using OpenCV to detect human objects and then applying gait recognition

Using the Neural Network as an Engine after detection of a walking human being, and there after using gait to recognize the identity of the person.

On executing detect_and_recognize.py it should access the default camera API and start looking for the object and then use the gait recognition model weights to recognize the person.