Realsee (beijing) Technology Co., Ltd.

Ke.com (beijing)technology Co., Ltd.

Tong Rao

Mingyuan Wang

Yonglin Yang

Yutong Chen

Wei Li

Yang Hu

Jie Zhou

Yi Zhu

Xiaodong Gu

Cihui Pan

Chenglin Liu

Chengfeng Huang

A method for image processing is disclosed herein. A computer system obtains a first image and a second image acquired by an imaging device. The computer system determines a matrix relating a first set of image points in the first image to a second set of image points in the second image. Based on the matrix, the computer system determines a set of parameters of the imaging device and a rotational relationship between the first image and the second image corresponding to poses of the imaging device. In response to determining that the first image and the second image are aligned based on the set of parameters of the imaging device and the rotational relationship between the first image and the second image, the computer system generates a composite image of the first image and the second image and causes display of the composite image.

Method, apparatus and system for image processing

Embodiments of the present disclosure disclose a method and device for vertical correction of a panoramic image, an electronic apparatus and a storage medium. The method includes extracting a set of straight line segments from the panoramic image, where an included angle between each straight line segment in the set of straight line segments and a current vertical direction of the panoramic image is less than a preset angle. The method further includes determining a set of target points by mapping the set of straight line segments on a unit sphere of the panoramic image. The method also includes determining an equatorial plane based on the set of target points and the center of the unit sphere and determining a vertical correction matrix based on the equatorial plane and a unit vector of the current vertical direction. The method additionally includes performing the vertical correction on the panoramic image using the vertical correction matrix.

Vertical correction for panoramic image

A method for generating a panoramic image is disclosed. The method includes receiving a three-dimensional model of a target space and an initial panoramic image of the target space, the initial panoramic image having a latitude span less than a preset latitude span; determining a first set of coordinate parameters of a camera associated with the initial panoramic image and with respect to a reference frame associated with the three-dimensional model; mapping, based on the first set of coordinate parameters of the camera, data points of the three-dimensional model to a camera coordinate system associated with the camera to obtain an intermediate panoramic image; and obtaining a final panoramic image by merging the initial panoramic image and the intermediate panoramic image, the final panoramic image having a latitude span greater than or equal to the preset latitude span.

Method and device for generating a panoramic image

A method for video processing is provided. The method comprises obtaining an image of a scene, obtaining a video that records an area included in the scene, determining one or more frames from the plurality of frames of the video, determining pairs of matched features, generating a plurality of composite frames by combining each of the selected one or more frames with the image of the scene based on the pairs of matched features, and generating a composite video based on the plurality of composite frames. Each of the pairs of matched features is related to an object that is in both the image and the one or more frames. Each of the pairs of matched features is associated with one or more pixels of the image of the scene and one or more pixels of a selected frame of the one or more frames.

Method, apparatus and system for video processing

A method for combining multiple images is disclosed herein. A target mapping matrix is determined based on a first image and a second image. The target mapping matrix is associated with a target correspondence between the first image and the second image. The first image and the second image are combined into a combined image based on the first target mapping matrix. The combined image is output by implementing the disclosed method.

Growing community of inventors

Beijing, China

Tong Rao