Arcsoft Corporation Limited

Hangzhou Fujia Gallium Technology Co. Ltd.

Zhejiang University

Alibaba (china) Co., Ltd.

Long Zhang

Wei Zhou

Wen Zhou

Hongzhuang Yang

Hui Deng

Wenxing Xi

Naiyang Lin

Duanyang Chen

Hongji Qi

Jin Wang

Ji Zhang

Feng Xie

Zhongzhou Zhao

Xinxing Zu

Zhiyong Xiao

The disclosed embodiments relate to a behavior control method and apparatus for a virtual live streaming character. In an embodiment, a method comprises: acquiring viewer behavior data of a live streaming service, and determining atmosphere data of the live streaming service according to the viewer behavior data; acquiring scenario data of the live streaming service, using the viewer behavior data, the atmosphere data, and the scenario data as behavior decision input data, and performing multi-modal feedback behavior decision-making based on the behavior decision input data to acquire a behavior control instruction; and performing a corresponding behavior operation according to the behavior control instruction. The disclosed embodiments combine existing knowledge and newly added behavior, achieves targeted behavior decision-making and control, improves the accuracy of virtual human behavior control, and considers both flexibility and comprehensiveness, thereby greatly improving the quality of virtual human behavior, and improving viewing experience of users.

Behavior control method and apparatus for virtual live streaming character

A preparation method of conductive gallium oxide based on deep learning and vertical Bridgman growth method. The prediction method includes: obtaining a preparation data of the conductive gallium oxide single crystal, the preparation data includes a seed crystal data, an environmental data, a control data and a raw material data, and the raw material data includes a doping type data and a conductive doping concentration; preprocessing the preparation data to obtain a preprocessed preparation data; inputting the preprocessed preparation data into a trained neural network model, and obtaining a predicted property data corresponding to the conductive gallium oxide single crystal through the trained neural network model, the predicted property data includes a predicted carrier concentration.

Preparation method of conductive gallium oxide based on deep learning and vertical Bridgman growth method

The present application discloses a preparation method of high resistance gallium oxide based on deep learning and vertical Bridgman growth method. The prediction method comprises: obtaining a preparation data of the high resistance gallium oxide single crystal, the preparation data comprises a seed crystal data, an environmental data, a control data and a raw material data, and the raw material data comprises a doping type data and a doping concentration; preprocessing the preparation data to obtain a preprocessed preparation data; inputting the preprocessed preparation data into a trained neural network model, and obtaining a predicted property data corresponding to the high resistance gallium oxide single crystal through the trained neural network model, the predicted property data comprises a predicted resistivity.

Preparation method of high resistance gallium oxide based on deep learning and vertical bridgman growth method

A method and a device for image processing are disclosed. The method includes: receiving a first image, wherein the first image includes a face; detecting the face and a background region in the first image, establishing a three-dimensional model of the face according to the first image; rotating the three-dimensional model of the face by a first angle; and projecting the three-dimensional model of the face rotated by the first angle to an image coordinate system of the first image and fusing a face region with a processed background region to obtain a second image.

Method and device for image processing

A method for image processing includes: receiving a third two-dimensional image and a depth image corresponding to the third two-dimensional image, wherein the third two-dimensional image and the depth image include a face; establishing a three-dimensional model of the face according to the depth image; rotating the three-dimensional model of the face by a first angle; projecting the three-dimensional model of the face rotated by the first angle to an image coordinate system of the third two-dimensional image; and building a three-dimensional model of a background region of the third two-dimensional image, processing a background region of an image projected to the image coordinate system of the third two-dimensional image to obtain a fourth image.

The present invention discloses an application of the exosome T&#x3b2;RII protein as a marker in preparing a breast cancer diagnostic kit, and belongs to the technology field of breast cancer detection kits. The present invention finds that the content of T&#x3b2;RII positive exosomes in the serum of breast cancer patients is significantly higher than these of healthy people, and after the surgery, the content of T&#x3b2;RII exosomes in the serum of breast cancer patients reduces significantly, which suggests the value of the T&#x3b2;RII positive exosomes during the diagnosis and prognosis of breast cancer. The present invention provides a breast cancer diagnosis kit based on detecting exosome T&#x3b2;RII protein in peripheral blood, which is simple in operation, and only needs to extract a small amount of peripheral blood of the test population, thereby detecting whether the test population has breast cancer and evaluating the tumor of the patient. The degree of malignancy and the state of metastasis.

Application of exosome T&#x3b2;RII protein as a marker in the preparation of breast cancer detection kit

The disclosure provides a method and device for three-dimensional reconstruction, applied to the field of image processing. The method includes: obtaining a first depth map, which is photographed by a first photographic device, and obtaining a second depth map, which is photographed by a second photographic device; merging the first depth map with a first three-dimensional model according to a position of the first photographic device to obtain a second three-dimensional model; and merging the second depth map with the second three-dimensional model according to a position of the second photographic device to obtain a third three-dimensional model.

Method and device for three-dimensional reconstruction

The present invention provides a method and a device for image processing, which are applied to the field of image processing. The method includes: receiving a first image, wherein the first image includes a face; detecting the face and a background region in the first image, establishing a three-dimensional model of the face according to the first image; rotating the three-dimensional model of the face by a first angle; projecting the three-dimensional model of the face rotated by the first angle to an image coordinate system of the first image, and fusing a face region with a processed background region to obtain a second image. The prevent invention provides a method and device for image processing, which can adjust the angle of a face in a captured image to achieve the effect of face enhancement.

The present invention provides a method and a device for three-dimensional modeling, used in the field of image processing. The method includes: receiving a first two-dimensional image and a depth map corresponding to the first two-dimensional image; the first two-dimensional image and the depth map respectively include a face; fitting three-dimensional face model by a first three-dimensional face database according to a position of a face feature point in the first two-dimensional image and the depth map corresponding to the first two-dimensional image. The method and the device for three-dimensional modeling provided by the present invention can improve the efficiency of the facial three-dimensional modeling and the precision of modeling.

Method and device for three-dimensional modeling

The present invention provides a method and device for three-dimensional reconstruction, applied to the field of image processing. The method comprises: obtaining a first depth map, which is photographed by a first photographic device, and obtaining a second depth map, which is photographed by a second photographic device; merging the first depth map with a first three-dimensional model according to a position of the first photographic device to obtain a second three-dimensional model; and merging the second depth map with the second three-dimensional model according to a position of the second photographic device to obtain a third three-dimensional model. By adoption of the method, during the reconstruction of three-dimensional images, the reconstruction effect of the three-dimensional images of the top and the bottom of the target object can be improved, and the precision of the reconstructed three-dimensional images is improved.

The present invention provides a method and device for calibration, applied to the field of image processing. The method comprises: obtaining a first two-dimensional image of a calibration object with a marking point and a first depth map of the calibration object, obtaining calibrated three-dimensional model data of the calibration object, determining a first three-dimensional marking point in the first depth map corresponding to the marking point based on the marking point in the first two-dimensional image, determining a second three-dimensional marking point in the three-dimensional model data of the calibration object corresponding to the marking point based on the marking point in the first two-dimensional image, and determining a first matrix. By adoption of the method mentioned above, when the device for photographing the three-dimensional image is calibrated, it is simple and feasible, and the operation difficulty is small.

Growing community of inventors

Hangzhou, China

Long Zhang