Abstract: Gesture recognition methods and systems are provided. First, a plurality of gesture templates are provided, wherein each gesture template defines a first gesture characteristic and a corresponding specific gesture. Then, a plurality of images is obtained, and a multi-background model is generated accordingly. At least one object image is obtained according to the multi-background model, wherein the object image includes at least an object having a plurality of edges. The included angles of any two adjacent edges of the object image are gathered as statistics to obtain a second gesture characteristic corresponding to the object image. The second gesture characteristic of the object image is compared with the first gesture characteristic of each gesture template. The specific gesture corresponding to the first gesture characteristic is obtained, when the second gesture characteristic is similar to the first gesture characteristic.

Abstract: An optical scanning-type touch apparatus and an operation method thereof are provided. The optical scanning-type touch apparatus includes a touch area, a light scanning module, an imaging module and a calculating unit. The light scanning module is disposed at one corner of the touch area to emit a scanning light for scanning the touch area. The imaging module is disposed at another corner of the touch area adjacent the light scanning module to obtain a first angle between the scattering light and a edge of the touch area, wherein the edge of the touch area is between the light scanning module and the imaging module. The calculating unit receives the first angle, a second angle and a distance between the light scanning module and the imaging module to calculate a position of the object on the touch area.

Abstract: A three-dimensional surround scanning device and a method thereof are described, which are adopted to perform surround scanning on a scene area, so as to construct a three-dimensional model. The device includes an image acquisition element, a first moving mechanism, a range acquisition element, and a controller. The controller controls the image acquisition element, the range acquisition element, and the first moving mechanism to perform three-dimensional image acquisition, so as to obtain a two-dimensional image covering the scene area, depth information with three-dimensional coordinates, and corresponding position signals. The controller rearranges and combines the two-dimensional image, position signals, and depth information, so as to construct the three-dimensional model.

Abstract: A computerized data equidistant-mapping method and a computer system using the same are provided. The computerized data equidistant-mapping method is used for mapping M sets of K-dimensional original data into a D-dimensional space in a manner of equidistance. The computerized data equidistant-mapping method includes the following steps. M sets of D-dimensional mapping data each corresponding to each original data respectively and sequentially are generated. An initial mapping data and an initial original data are chosen arbitrarily, wherein the initial mapping data corresponds to the initial original data. The mapping distance between each mapping data and the initial mapping data is computed. The original distance between each original data and the initial original data is computed. Each mapping data is amended according to each mapping distance and each original distance for enabling each mapping data and each original data to have equidistant-mapping relationship gradually.

Abstract: Gesture recognition methods and systems are provided. First, a plurality of gesture templates are provided, wherein each gesture template defines a first gesture characteristic and a corresponding specific gesture. Then, a plurality of images is obtained, and a multi-background model is generated accordingly. At least one object image is obtained according to the multi-background model, wherein the object image includes at least an object having a plurality of edges. The included angles of any two adjacent edges of the object image are gathered as statistics to obtain a second gesture characteristic corresponding to the object image. The second gesture characteristic of the object image is compared with the first gesture characteristic of each gesture template. The specific gesture corresponding to the first gesture characteristic is obtained, when the second gesture characteristic is similar to the first gesture characteristic.

Abstract: The present invention provides a multi-touch position tracking technique and an interactive system and a multi-touch interactive image processing method using the same. In the present invention, a light guide element is designed to comprise frustrating structures to frustrate total internal reflection (TIR) so that the light beam therein can be dispersed to form a dispersed optical field distribution. The dispersed optical field is used to respond a physical relation between an object and the light guide element.

Abstract: A three-dimensional surround scanning device and a method thereof are described, which are adopted to perform surround scanning on a scene area, so as to construct a three-dimensional model. The device includes an image acquisition element, a first moving mechanism, a range acquisition element, and a controller. The controller controls the image acquisition element, the range acquisition element, and the first moving mechanism to perform three-dimensional image acquisition, so as to obtain a two-dimensional image covering the scene area, depth information with three-dimensional coordinates, and corresponding position signals. The controller rearranges and combines the two-dimensional image, position signals, and depth information, so as to construct the three-dimensional model.

Abstract: A computerized data equidistant-mapping method and a computer system using the same are provided. The computerized data equidistant-mapping method is used for mapping M sets of K-dimensional original data into a D-dimensional space in a manner of equidistance. The computerized data equidistant-mapping method includes the following steps. M sets of D-dimensional mapping data each corresponding to each original data respectively and sequentially are generated. An initial mapping data and an initial original data are chosen arbitrarily, wherein the initial mapping data corresponds to the initial original data. The mapping distance between each mapping data and the initial mapping data is computed. The original distance between each original data and the initial original data is computed. Each mapping data is amended according to each mapping distance and each original distance for enabling each mapping data and each original data to have equidistant-mapping relationship gradually.

Abstract: A method that converts high-level scripts into 3D animations uses natural language for fast prototyping. The proposed method first transforms the natural language script into the metadata with high-level semantics, then matches the metadata to several motion clips in the motion databases, and finally, synthesizes these motion clips into a 3D animation.

Abstract: A method that converts high-level scripts into 3D animations uses natural language for fast prototyping. The proposed method first transforms the natural language script into the metadata with high-level semantics, then matches the metadata to several motion clips in the motion databases, and finally, synthesizes these motion clips into a 3D animation.