Lemon Inc.

Beijing Zitiao Network Technology Co., Ltd.

Shen Sang

Jing Liu

Chunpong Lai

Xu Wang

Weihong Zeng

Jingna Sun

Linjie Luo

Guoxian Song

Tiancheng Zhi

Peibin Chen

A method of generating a stylized 3D avatar is provided. The method includes receiving an input image of a user, generating, using a generative adversarial network (GAN) generator, a stylized image, based on the input image, and providing the stylized image to a first model to generate a first plurality of parameters. The first plurality of parameters include a discrete parameter and a continuous parameter. The method further includes providing the stylized image and the first plurality of parameters to a second model that is trained to generate an avatar image, receiving, from the second model, the avatar image, comparing the stylized image to the avatar image, based on a loss function, to determine an error, updating the first model to generate a second plurality of parameters that correspond to the first plurality of parameters, based on the error, and providing the second plurality of parameters as an output.

Cascaded domain bridging for image generation

A method of generating a style image is described. The method includes receiving an input image of a subject. The method further includes encoding the input image using a first encoder of a generative adversarial network (GAN) to obtain a first latent code. The method further includes decoding the first latent code using a first decoder of the GAN to obtain a normalized style image of the subject, wherein the GAN is trained using a loss function according to semantic regions of the input image and the normalized style image.

Systems and methods for rendering a translucent object are provided. In one aspect, the system includes a processor coupled to a storage medium that stores instructions, which, upon execution by the processor, cause the processor to receive at least one mesh representing at least one translucent object. For each pixel to be rendered, the processor performs a rasterization-based differentiable rendering of the pixel to be rendered using the at least one mesh and determines a plurality of values for the pixel to be rendered based on the rasterization-based differentiable rendering. The rasterization-based differentiable rendering can include performing a probabilistic rasterization process along with aggregation techniques to compute the plurality of values for the pixel to be rendered. The plurality of values includes a set of color channel values and an opacity channel value. Once values are determined for all pixels, an image can be rendered.

Methods for a rasterization-based differentiable renderer for translucent objects

The present disclosure provides a training method and device for an image identifying model, and an image identifying method. The training method comprises: obtaining image samples of a plurality of categories; inputting image samples of each category into a feature extraction layer of the image identifying model to extract a feature vector of each image sample; calculating a statistical characteristic information of an actual distribution function corresponding to each category according to the feature vector of each image sample of the each category; establishing an augmented distribution function corresponding to the each category according to the statistical characteristic information; obtaining augmented sample features of the each category based on the augmented distribution function; and inputting feature vectors of the image samples and the augmented sample features into a classification layer of the image identifying model for supervised learning.

Training method and device for image identifying model, and image identifying method

A method is provided for evaluating an effect of classifying a fuzzy attribute of an object, the fuzzy attribute referring to an attribute, a boundary between two similar ones of a plurality of categories of which is blurred, wherein the method includes: generating a similarity-based ranked confusion matrix, which comprises: based on similarities of K categories of the fuzzy attribute of the object, ranking the K categories, where K is an integer greater than or equal to 2, generating a K×K all-zero initialization matrix, wherein an abscissa and an ordinate of the initialization matrix respectively represent predicted values and true values of the similarity-based ranked categories of the fuzzy attribute, and based on the true values and the predicted values of the category of the fuzzy attribute for the multiple object samples, updating values of corresponding elements in the initialization matrix; and displaying the similarity-based ranked confusion matrix.

Method and device for evaluating effect of classifying fuzzy attribute

An image processing method includes acquiring a set of image samples for training an attribute recognition model, wherein the set of image samples includes a first subset of image samples with category labels and a second subset of image samples without category labels; training a sample prediction model using the first subset of image samples, and predicting categories of the image samples in the second subset of image samples using the trained sample prediction model; determining a category distribution of the set of image samples based on the category labels of the first subset of image samples and the predicted categories of the second subset of image samples; and acquiring a new image sample if the determined category distribution does not conform to the expected category distribution, and adding the acquired new image sample to the set of image samples.

Image processing method, image processing device and computer readable medium, for acquiring image sample data for training an attribute recognition model

Growing community of inventors

Los Angeles, CA, United States of America

Shen Sang