Abstract: A camera and method identify moving objects of interest in a field of view of the camera. The method includes: capturing two or more images successively over a time period, each image being associated with different times during the time period; obtaining binary image from each successive pair of images, the binary image comprising a binary value at each pixel indicating whether a change in pixel values of at least a predetermined magnitude has occurred at that pixel between the time associated with the first image of the success pair of images and time associated with the second image of the successive pair of images; deriving one or more motion boxes each encapsulating one or more nearby pixels in binary image; processing motion boxes of each binary image to obtain refined motion boxes; and classifying refined motion boxes each into categories representative of one moving object of interest.

Abstract: Embodiments of the present disclosure include a non-transitory computer-readable medium with computer-executable instructions stored thereon executed by one or more processors to perform a method to select and implement a neural network for an embedded system. The method includes selecting a neural network from a library of neural networks based on one or more parameters of the embedded system, the one or more parameters constraining the selection of the neural network. The method also includes training the neural network using a dataset. The method further includes compressing the neural network for implementation on the embedded system, wherein compressing the neural network comprises adjusting at least one float of the neural network.