Htc Corporation

Institute for Information Industry

Li-Cheng Chen

Hsin-Ti Chueh

Huai-Che Lee

Yung-Chao Tseng

Jhih-Rong Chen

Jing-Lung Wu

Jenn-Wein Wu

Pol-Lin Tai

Chun-Fa Chang

Hsiang-Yun Chang

A method for creating a parallax video from a still image includes following steps. Image segmentation is performed for separating at least a foreground object within the still image from an image background of the still image. A blank region of the image background is filled, and the blank region corresponds to where the separated foreground object locates within the still image. A virtual three-dimensional (3D) scene consisting of the foreground object and the image background is created. The virtual 3D scene is analyzed for determining a camera path; and, creating the parallax video by observing the virtual 3D scene with a virtual camera moving along the camera path.

Method of creating a parallax video from a still image

An apparatus and computer-implemented method for generating a three-dimensional scene and non-transitory tangible computer readable medium thereof are provided. The apparatus comprises a storage unit and a processing unit. The storage unit is stored with an image and a plurality of depth data of the image. The processing unit is configured to generate a depth histogram according to the depth data, generate a region of interest of the image according to the depth histogram, decide a moving path of a virtual camera according to the region of interest, and generate the three-dimensional scene according to the image and the moving path.

Apparatus and computer-implemented method for generating a three-dimensional scene and non-transitory tangible computer readable medium thereof

An image processing method includes steps of: providing a source face image and a target face image; extracting facial features from the source face image and the target face image respectively; detecting feature dimensions of the facial features from the source face image and the target face image respectively; pairing the facial features from the source face image with the facial features from the target face image; and, forming an output face image by adjusting the facial features from the source face image in at least one of the feature dimensions according to the paired features from the target face image in the corresponding feature dimensions.

Image processing with facial features

Apparatuses, methods, and non-transitory tangible computer readable media thereof for creating a 3D scene are provided. The apparatus generates a height map according to a plurality of depth data of an image. The apparatus finds a first region of the height map, wherein the depth data within the first region change more greatly than the depth data outside the first region. The apparatus creates a plurality of grids on a plane according to the first region and generate a 3D mesh by morphing the height map with the grids of the plane. The plane and the height map are of the same size, a second region within the plane corresponds to the first region of the height map, and the grids inside the second region has a finer resolution than the grids outside the second region. The apparatus generates the 3D scene by mapping the image onto the 3D mesh.

Apparatus, method, and non-transitory tangible computer readable medium thereof for creating 3D scene

The invention discloses an image cropping manipulation method for use in a portable electronic device. The portable electronic device comprises a touch screen and an image processing unit. The method comprises: displaying a sub-image on the touch screen, the sub-image is cropped from a digital image; receiving a user input from the touch screen; in response to the user input, map the user input to an operation on the digital image; and performing the operation to update the sub-image.

Image cropping manipulation method and portable electronic device

A method for rendering fluid is provided. First, state information of a plurality of fluid particles is provided, wherein the state information records whether the fluid particles are located above or under a fluid surface and the interactions between the fluid particles and a terrain or the dynamic objects. Then, whether to render the fluid particles in a direction facing a viewer or in a direction parallel to the flow direction is determined according to the information that whether the fluid particles are located above or under the fluid surface. Next, the fluid particles are rendered as a plurality of two-dimensional metaballs according to the interactions between the fluid particles and the terrain or the dynamic objects, and these metaballs are stacked to reconstruct the fluid.

Growing community of inventors

Taoyuan County, Taiwan

Li-Cheng Chen