Beijing Dajia Internet Information Technology Co., Ltd.

Wen Zheng

Yufeng Liu

Lei Zhang

Wenbo Zhang

Congli Song

Hongmin Xu

Huijuan Huang

Qiang Li

Yuan Gao

Yang Sun

Songtao Zhao

Yilin Guo

Zairan Wang

Xiaoyan Guo

Jili Gu

Yukun Hu

Yajiao Dong

A method for transforming hair relates to the field of image processing, and can achieve a function of transforming the hair with a more consistent and natural color with respect to that before the hair of a character is transformed. The method for transforming the hair includes: acquiring a first sample image and a second sample image; determining a first color guide map based on the first sample image and the second sample image; acquiring a first image processing model by training with the first sample image and the first color guide map as training data and the second sample image as supervision information; and transforming, based on the first image processing model, hair in an image selected by a user.

Method and apparatus for transforming hair

A method for detecting gesture key points can include: acquiring a target image to be detected; determining a gesture category according to the target image, the gesture category being a category of a gesture contained in the target image; determining a target key point detection model corresponding to the gesture category from a plurality of key point detection models; and performing a key point detection on the target image by the target key point detection model.

Method and device for detecting hand gesture key points

The disclosure can provide a method for converting an image, an apparatus for converting an image, an electronic device, and a storage medium. The method includes: obtaining an image in RAW format; obtaining a semantic analysis result of the image in RAW format by inputting the image in RAW format into a pre-trained first network model; the first network model being obtained by training based on a labeled training sample corresponding to each of a plurality of first training samples, the first training sample being a sample image in RAW format, and the labeled training sample being a sample image in RAW format obtained by labeling a semantic analysis result on the corresponding first training sample; and determining an image in RGB (Red-Green-Blue) format corresponding to the image in RAW format based on the semantic analysis result of the image in RAW format.

Method, device, and storage medium for converting image from raw format to RGB format

A method for detecting key points of skeleton, an apparatus, an electronic device, and a storage medium are provided. The method is implemented as follows. An original image is acquired. The original image includes a plurality of key points of skeleton. Based on a pre-trained stacked hourglass network structure, skeleton key point identification is performed on the original image to obtain heat maps of the plurality of key points. The stacked hourglass network structure includes at least one hourglass network. The at least one hourglass network is configured to perform deep-layer feature learning on feature maps of the plurality of key points based on weight values corresponding to the feature maps.

Method for detecting key points in skeleton, apparatus, electronic device and storage medium

The present disclosure relates to a method, a device and a medium for making up a face. The method for making up the face of the present disclosure includes: obtaining a first face image; determining facial key-points by detecting the first face image; generating a second face image by applying makeup to a face in the first face image based on the facial key-points; determining a first face region by segmenting the first face image, wherein the first face region is a face region that is not shielded in the first face image; and generating a final face makeup image with makeup based on the first face region and the second face image.

Method, device, and medium for processing image

A method and apparatus for detecting a moving target, an electronic equipment and a storage medium. The method includes obtaining a first frame image and a second frame image which are adjacent, and a rotation matrix and a translation matrix between the first and second frame images, the first and second frame image include the same moving target; extracting first feature points from the first frame image; determining second feature points corresponding to the first feature points from the second frame image based on the second frame image and the first feature points; determining distances between the second feature points and corresponding polar lines based on the rotation matrix and the translation matrix; determining third feature points located on the moving target based on the distances; and detecting the moving target based on the third feature points.

Method and apparatus for detecting moving target, and electronic equipment

Disclosed are a method, an electronic device, and a storage medium of gesture recognition. The method includes: acquiring a hand image; extracting a first standard feature of a hand in the hand image based on a feature mapping model; the feature mapping model being obtained by training based on second standard features of hands in a synthesized image sample and a real image sample; obtaining three-dimensional coordinates of multiple key points of the hand by processing the first standard feature; determining a gesture of the hand based on the three-dimensional coordinates of the multiple key points

Method and electronic device of gesture recognition

A method for detecting a posture, including: acquiring an image to be detected; and inputting the image to be detected into a pre-trained detection model, and detecting a posture category in the image to be detected. The detection model is acquired by training based on a positive sample image and a negative sample image, a basic posture categorization model is acquired by training based on the positive sample image with a marked preset posture category, and the negative sample image is determined based on the basic posture categorization model, and the negative sample image contains no posture of the preset posture category and the positive sample image is an image inputted by an user or an image stored at a preset storage location.

Posture detection method, apparatus and device, and storage medium

The application shows a method, an apparatus, and an electronic device for processing an image. The method includes: obtaining an image of a first rendering style; obtaining an image of a second rendering style according to the image of the first rendering style and a first preset processing model; generating at least one first intermediate gradient image according to the image of the first rendering style, the image of the second rendering style, and a second preset processing model, the at least one first intermediate gradient image comprising an image in a process of gradient from the image of the first rendering style to the image of the second rendering style; generating a first gradient video according to the image of the first rendering style, at least one first intermediate gradient image, and an image of the second rendering style.

Growing community of inventors

Beijing, China

Wen Zheng