The 'Vision' submodule will emulate human vision. It will allows robots to percieve as humans do.
At the moment, this project is a real-time object recognition application. It does so by using Google's TensorFlow Object Detection API and OpenCV.
The detected objects are classified using labelled bounding boxes, but a tracker will save the corresponding pixels in a dynamic memory for the 'Perception' and 'Memory' module.
In the future:
Vision: Detection - Recognition
When it detects a face: -then recognize identity -sentiment analysis: -Blink Detection: blink rate -visual attention detection (gaze tracking): -position: face sideways, tilted ? -age ? -describe image ?
python object_detection_app.py
Optional arguments (default value):
- Device index of the camera
--source=0 - Width of the frames in the video stream
--width=480 - Height of the frames in the video stream
--height=360 - Number of workers
--num-workers=2 - Size of the queue
--queue-size=5
pytest -vs utils/
Using Windows x64 at the moment of deployment.
Object Detection: -https://github.com/datitran/object_detector_app -See LICENSE for details. -Copyright (c) 2017 Dat Tran.