Motion-based object detection method, object detection apparatus and electronic device

An object detection apparatus, comprising:

• a region of interest (ROI) extractor configured to extract, by processing acquired first and second images, one or more ROIs;
• a grayscale transformer configured to transform the one or more ROIs into grayscale;
• and
• a classification result acquirer configured to acquire, by processing the grayscale ROIs with a deep neural network (DNN) model to classify objects contained in the one or more ROIs, a classification result of whether the objects contained in the one or more ROIs belong to a given categories, wherein the DNN model comprises N depthwise separable convolution layers, wherein each depthwise separable convolution layer comprises a depthwise convolution layer for applying a single filter to each input channel and a pointwise layer for linearly combining outputs of the depthwise convolution layer to obtain feature maps of the grayscale ROIs, wherein N is a positive integer and ranged from 4-12.

A non-transitory computer storage medium that, when executed by a processor, causes the processor to perform the following method:

• processing acquired first and second images, one or more ROIs;
• transforming the one or more ROIs into grayscale;
• and
• acquiring, by processing the grayscale ROIs with a deep neural network (DNN) model to classify objects contained in the one or more ROIs, a classification result of whether the objects contained in the one or more ROIs belong to a given categories, wherein the DNN model comprises N depthwise separable convolution layers, wherein each depthwise separable convolution layer comprises a depthwise convolution layer for applying a single filter to each input channel and a pointwise layer for linearly combining outputs of the depthwise convolution layer to obtain feature maps of the grayscale ROIs, wherein N is a positive integer and ranged from 4-12.