公开(公告)号：US20240265621A1

公开(公告)日：2024-08-08

申请号：US18165794

申请日：2023-02-07

Applicant: Lemon Inc.

Inventor： Hongyi Xu ,  Guoxian Song ,  Zihang Jiang ,  Jianfeng Zhang ,  Yichun Shi ,  Jing Liu ,  Wanchun Ma ,  Jiashi Feng ,  Linjie Luo

IPC: G06T15/08 ,  G06T3/40 ,  G06V40/16

CPC classification number: G06T15/08 ,  G06T3/4046 ,  G06T3/4053 ,  G06V40/176

Abstract: Technologies are described and recited herein for producing controllable synthesized images include a geometry guided 3D GAN framework for high-quality 3D head synthesis with full control on camera poses, facial expressions, head shape, articulated neck and jaw poses; and a semantic SDF (signed distance function) formulation that defines volumetric correspondence from observation space to canonical space, allowing full disentanglement of control parameters in 3D GAN training.

公开(公告)号：US20240135627A1

公开(公告)日：2024-04-25

申请号：US18046077

申请日：2022-10-12

Applicant: Lemon INc.

Inventor： Guoxian SONG ,  Shen Sang ,  Tiancheng Zhi ,  Jing Liu ,  Linjie Luo

IPC: G06T15/02 ,  G06T7/11

CPC classification number: G06T15/02 ,  G06T7/11 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30201

Abstract: 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.

公开(公告)号：US20240078792A1

公开(公告)日：2024-03-07

申请号：US17929449

申请日：2022-09-02

Applicant: Lemon Inc.

Inventor： Shuo Cheng ,  Wanchun Ma ,  Linjie Luo

IPC: G06V10/774 ,  G06V10/764 ,  G06V10/776 ,  G06V10/82 ,  G06V10/96 ,  G06V40/16

CPC classification number: G06V10/774 ,  G06V10/764 ,  G06V10/776 ,  G06V10/82 ,  G06V10/96 ,  G06V40/171 ,  G06V40/174

Abstract: Systems and methods for multi-task joint training of a neural network including an encoder module and a multi-headed attention mechanism are provided. In one aspect, the system includes a processor configured to receive input data including a first set of labels and a second set of labels. Using the encoder module, features are extracted from the input data. Using a multi-headed attention mechanism, training loss metrics are computed. A first training loss metric is computed using the extracted features and the first set of labels, and a second training loss metric is computed using the extracted features and the second set of labels. A first mask is applied to filter the first training loss metric, and a second mask is applied to filter the second training loss metric. A final training loss metric is computed based on the filtered first and second training loss metrics.

公开(公告)号：US11720994B2

公开(公告)日：2023-08-08

申请号：US17321384

申请日：2021-05-14

Applicant: Lemon Inc.

Inventor： Linjie Luo ,  Guoxian Song ,  Jing Liu ,  Wanchun Ma

IPC: G06T3/00 ,  G06T11/00 ,  G06T5/00 ,  G06N3/08 ,  G06F18/214 ,  G06N3/045

CPC classification number: G06T3/0012 ,  G06F18/214 ,  G06N3/045 ,  G06N3/08 ,  G06T3/0006 ,  G06T5/00 ,  G06T11/00 ,  G06T2207/20016 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30201

Abstract: Systems and method directed to an inversion-consistent transfer learning framework for generating portrait stylization using only limited exemplars. In examples, an input image is received and encoded using a variational autoencoder to generate a latent vector. The latent vector may be provided to a generative adversarial network (GAN) generator to generate a stylized image. In examples, the variational autoencoder is trained using a plurality of images while keeping the weights of a pre-trained GAN generator fixed, where the pre-trained GAN generator acts as a decoder for the encoder. In other examples, a multi-path attribute aware generator is trained using a plurality of exemplar images and learning transfer using the pre-trained GAN generator.

公开(公告)号：US20230124252A1

公开(公告)日：2023-04-20

申请号：US17501990

申请日：2021-10-14

Applicant: Lemon Inc.

Inventor： Jing Liu ,  Chunpong Lai ,  Guoxian Song ,  Linjie Luo ,  Ye Yuan

IPC: G06T5/00 ,  G06T3/40 ,  G06T5/50 ,  G06T11/00 ,  G06T7/11

Abstract: Systems and method directed to generating a stylized image are disclosed. In particular, the method includes, in a first data path, (a) applying first stylization to an input image and (b) applying enlargement to the stylized image from (a). The method also includes, in a second data path, (c) applying segmentation to the input image to identify a face region of the input image and generate a mask image, and (d) applying second stylization to an entirety of the input image and inpainting to the identified face region of the stylized image. Machine-assisted blending is performed based on (1) the stylized image after the enlargement from the first data path, (2) the inpainted image from the second data path, and (3) the mask image, in order to obtain a final stylized image.