-
公开(公告)号：US20190080148A1
公开(公告)日：2019-03-14
申请号：US16050329
申请日：2018-07-31
发明人： Tao HE , Gang Zhang , Jingtuo Liu
摘要： The present disclosure discloses a method and apparatus for generating an image. A specific embodiment of the method comprises: acquiring at least two frames of facial images extracted from a target video; and inputting the at least two frames of facial images into a pre-trained generative model to generate a single facial image. The generative model updates a model parameter using a loss function in a training process, and the loss function is determined based on a probability of the single facial generative image being a real facial image and a similarity between the single facial generative image and a standard facial image. According to this embodiment, authenticity of the single facial image generated by the generative model may be enhanced, and then a quality of a facial image obtained based on the video is improved.
-
公开(公告)号：US20180204052A1
公开(公告)日：2018-07-19
申请号：US15533939
申请日：2015-11-27
发明人： Guang Li , Jiawei Gu , Shasha Yan , Jianping Hao , Jiabin He , Li Xu , Jingtuo Liu , Fuguo Zhu , Zhengping Wei
IPC分类号： G06K9/00
摘要： The present disclosure discloses a method and apparatus for human face image processing. A specific embodiment of the method comprises: locating facial feature points in a human face image, extracting an image of a human face region according to a range defined by the facial feature points, transforming a facial image of the source image according to a face shape of the target image, transforming the facial image of the source image according to a complexion distribution of a facial region of the target image, and obtaining a new human face by combining the facial image of the source image and a facial image of the target image. The embodiment achieves a facial image processing with higher similarity to the user in the image, of simple steps, small calculation and high real-time performance.
-
公开(公告)号：US10599914B2
公开(公告)日：2020-03-24
申请号：US15533939
申请日：2015-11-27
发明人： Guang Li , Jiawei Gu , Shasha Yan , Jianping Hao , Jiabin He , Li Xu , Jingtuo Liu , Fuguo Zhu , Zhengping Wei
摘要： The present disclosure discloses a method and apparatus for human face image processing. A specific embodiment of the method comprises: locating facial feature points in a human face image, extracting an image of a human face region according to a range defined by the facial feature points, transforming a facial image of the source image according to a face shape of the target image, transforming the facial image of the source image according to a complexion distribution of a facial region of the target image, and obtaining a new human face by combining the facial image of the source image and a facial image of the target image. The embodiment achieves a facial image processing with higher similarity to the user in the image, of simple steps, small calculation and high real-time performance.
-
公开(公告)号：US11978245B2
公开(公告)日：2024-05-07
申请号：US16050329
申请日：2018-07-31
发明人： Tao He , Gang Zhang , Jingtuo Liu
IPC分类号： G06V10/82 , G06F18/214 , G06F18/2413 , G06N3/044 , G06N3/045 , G06N3/047 , G06N3/08 , G06N3/084 , G06N7/00 , G06N7/01 , G06N20/00 , G06N20/10 , G06T7/00 , G06T7/246 , G06T11/00 , G06V10/764 , G06V40/16
CPC分类号： G06V10/82 , G06F18/214 , G06F18/2413 , G06N3/044 , G06N3/045 , G06N3/047 , G06N3/08 , G06N3/084 , G06N7/00 , G06N20/00 , G06T7/251 , G06T7/97 , G06T11/00 , G06V10/764 , G06V40/165 , G06V40/167 , G06V40/168 , G06V40/172 , G06N7/01 , G06N20/10 , G06T2207/20081 , G06T2207/30201
摘要： The present disclosure discloses a method and apparatus for generating an image. A specific embodiment of the method comprises: acquiring at least two frames of facial images extracted from a target video; and inputting the at least two frames of facial images into a pre-trained generative model to generate a single facial image. The generative model updates a model parameter using a loss function in a training process, and the loss function is determined based on a probability of the single facial generative image being a real facial image and a similarity between the single facial generative image and a standard facial image. According to this embodiment, authenticity of the single facial image generated by the generative model may be enhanced, and then a quality of a facial image obtained based on the video is improved.
-
公开(公告)号：US10860839B2
公开(公告)日：2020-12-08
申请号：US16103122
申请日：2018-08-14
发明人： Renliang Weng , Tao He , Gang Zhang , Jingtuo Liu , Wenxian Liu
IPC分类号： G06K9/00 , G06N3/04 , G06K9/62 , G06F17/15 , G06K9/66 , G06N20/00 , G06N3/08 , G06N7/00 , G06N20/10
摘要： The present disclosure provides a method and apparatus for generating training data for human face recognition, a device and a computer storage medium, wherein the method comprises: inputting accessory-not-worn face images into a generative network, to obtain accessory-worn face images; using the accessory-worn face images as second training data for building the human face recognition model; wherein the generative network is a generative network in a generative adversarial network obtained by pre-training with first training data, the first training data including the accessory-not-worn face images and accessory-worn face images corresponding to a same user identifier. In the present disclosure, the accessory-worn face images obtained in a data augmentation manner greatly expand the amount of training data for building the human face recognition model, and thereby improve the recognition accuracy of the accessory-worn face images.
-
公开(公告)号：US10650492B2
公开(公告)日：2020-05-12
申请号：US16052429
申请日：2018-08-01
发明人： Tao He , Gang Zhang , Jingtuo Liu
摘要： A method and an apparatus for generating an image are provided. A specific embodiment of the method comprises: acquiring a to-be-processed facial image, the image resolution of the to-be-processed facial image being lower than a preset first resolution threshold; and inputting the to-be-processed facial image into a pre-trained generativemodel to generate a processed facial image. The generative model updates a model parameter using a loss function in a training process, and the loss function is determined based on a probability of an image group being positive sample data, the image group composed of a facial sample image and a facial generative image. According to this embodiment, authenticity of the generated facial image is enhanced.
-
7.发明申请公开(公告)号：US20190050632A1
公开(公告)日：2019-02-14
申请号：US16103122
申请日：2018-08-14
发明人： Renliang Weng , Tao He , Gang Zhang , Jingtuo Liu , Wenxian Liu
摘要： The present disclosure provides a method and apparatus for generating training data for human face recognition, a device and a computer storage medium, wherein the method comprises: inputting accessory-not-worn face images into a generative network, to obtain accessory-worn face images; using the accessory-worn face images as second training data for building the human face recognition model; wherein the generative network is a generative network in a generative adversarial network obtained by pre-training with first training data, the first training data including the accessory-not-worn face images and accessory-worn face images corresponding to a same user identifier. In the present disclosure, the accessory-worn face images obtained in a data augmentation manner greatly expand the amount of training data for building the human face recognition model, and thereby improve the recognition accuracy of the accessory-worn face images.
-
-
-
-
-
-
搜索结果
7 条记录 (耗时 0.019 秒)
