Abstract: An image tracking method of the present invention includes the following steps: (A) obtaining a plurality of original images by using an image capturing device; (B) transmitting the plurality of original images to a computing device, and generating a position box based on a preset image set; (C) obtaining an initial foreground image including a target object, and an identified foreground image is determined based on a pixel ratio and a first threshold; (D) obtaining a feature and obtaining a first feature score based on the feature of the identified foreground images; and (E) generating a target object matching result based on the first feature score and a second threshold, and recording a moving trajectory of the target object based on the target object matching result.

Abstract: An image processing apparatus includes a low-resolution image generating circuit configured to generate a low-resolution image including a second pixel corresponding to first pixels based on an input image including the first pixels, and an edge preserving smoothing circuit configured to generate a reliability of the second pixel based on characteristics of values of the first pixels and perform edge preserving smoothing on the input image using a value of the second pixel of which a reflection ratio is adjusted, based on the reliability of the second pixel.

Abstract: An information processing device includes: a processor configured to: obtain sensor data; determine partial data; input the sensor data to an input layer of a first neural network recognition model; input the partial data to an input layer of a second neural network recognition model; align second intermediate output data from a second intermediate layer of the second neural network recognition model, relative to first intermediate output data from a first intermediate layer of the first neural network recognition model; generate intermediate input data from the first intermediate output data and the second intermediate output data aligned; and input the intermediate input data to a third intermediate layer of the first neural network recognition model subsequent to the first intermediate layer.