Industrial Technology Research Institute

National Yang Ming Chiao Tung University

Chih-Chung Kao

Chao-Jung Chen

Bing-Fei Wu

Meng-Liang Chung

Chung-Cheng Chiu

Hao-Ping Kang

Yao-Chin Yang

Min-Yu Ku

Chih-Chun Liu

Cheng-Yen Yang

Jia-Hua Wu

Ming-Hung Cheng

Wei-Cheng Sun

Ying-Han Chen

Sänket N⋅ Yerule

A method for detecting a human skeleton is provided. The method includes: receiving a video frame, wherein the video frame comprises a human body; determining whether the video frame comprises prediction information; determining whether a first intra-coded macroblock (IMB) ratio of a target area comprising the human body in the video frame is greater than a first threshold when the video frame comprises the prediction information; and using a motion vector (MV) to estimate skeleton information of the human body when the first IMB ratio of the target area is not greater than the first threshold.

Method and device for detecting human skeletons

A person re-identification method, a person re-identification system and an image screening method are provided. The person re-identification method includes the following steps. A plurality of identified images of an identified person are analyzed to obtain an interrelated array, which records a plurality of degrees of association among the identified images. The interrelated array is converted to a directed graph having at least one connected group. A representative path of each connected group is obtained. According to each representative path, a part of the identified images are outputted as at least one representative image set for performing a person re-identification process on an image to be identified.

Person re-identification method, person re-identification system and image screening method

A method for recognizing face expressions is provided. The method includes: recognizing expression categories of expressions in a plurality of face images and obtaining recognition results between each expression category and another expression category; obtaining similarities between each expression category and another expression category according to the recognition results; classifying the expression categories into a plurality of expression groups according to the similarities; training a first recognition model to classify the expressions in the face images into the expression groups; and training a second recognition model for each of the expression groups to classify the face images in each of the expression group into one of the expression categories.

Method and device for recognizing facial expressions

A character recognition method is provided, including the steps of: recognizing a character string, which includes one or more characters, in a region of interest (ROI) using a first convolutional neural network (CNN) model to generate a recognized character string; and comparing the recognized character string with a plurality of character strings in a character database to find a character string corresponding to the recognized character string, wherein each of the character strings includes one or more characters; wherein if the character string corresponding to the recognized character string is found, the recognized character string is used as a character recognition result, otherwise, using a second CNN model to correct the recognized string and generate a new recognized character string as the character recognition result.

Character recognition systems and character recognition methods thereof using convolutional neural network

The invention relates to a method for image processing, which can be used to calibrate the background quickly. When the external environment is changed due to the switch of light, the color of background is calibrated quickly, and the background can be updated together. The method not only is used to update the background, but also can be used to eliminate the convergence of background again.

Method for tracking and processing image

A parking assistance system (PAS) comprising an image capture device, a display device and an image analysis device is provided, wherein the image capture device captures a rear parking image of a car and the image analysis device analyses the parking image to determine an initial line, two side lines and an auxiliary line in the parking image. The image analysis device determines a first reference point according to a specific position on the car, a second reference point according to an interval of the initial line between the two side lines, a third reference point according to an interval of the auxiliary line between the two side lines and a recommended parking path according to these reference points, and transmits the parking image and the recommended parking path to the display device, such that the parking image and the recommended parking path are superimposed to be displayed on the display.

Parking assistance system and method

The invention relates to a method for image processing. First, establish the initial image background information. And retrieve the instant image information. Then calculate the initial image background information and color intensity information of the instant image. Furthermore, adjust the instant image information. Then calculate the moving-object information. Finally, track the moving-object information. It can improve the accuracy rate of detection without the influence of erected height.

Growing community of inventors

Pingtung County, Taiwan

Chih-Chung Kao