Patents by Inventor Shaoxiong YANG
Shaoxiong YANG has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11948236Abstract: The present disclosure discloses a method and apparatus for generating animation. An implementation of the method may include: processing a to-be-processed material to generate a normalized text; analyzing the normalized text to generate a Chinese pinyin sequence of the normalized text; generating a reference audio based on the to-be-processed material; and obtaining a animation of facial expressions corresponding to the timing sequence of the reference audio based on the Chinese pinyin sequence and the reference audio.Type: GrantFiled: November 16, 2021Date of Patent: April 2, 2024Assignee: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.Inventors: Shaoxiong Yang, Yang Zhao, Chen Zhao
-
Patent number: 11836836Abstract: Methods and apparatuses for generating a model and generating a 3D animation, devices, and storage mediums are provided. The method for generating a model may include: acquiring a preset sample set; acquiring pre-established generative adversarial nets, the generative adversarial nets including a generator and a discriminator; and performing training steps as follows: selecting a sample from the sample set; extracting a sample audio feature from the sample audio of the sample; inputting the sample audio feature into the generator to obtain a pseudo 3D mesh vertex sequence of the sample; inputting the pseudo 3D mesh vertex sequence and the real 3D mesh vertex sequence of the sample into the discriminator to discriminate authenticity of 3D mesh vertices; and in response to determining that the generative adversarial nets meet a training completion condition, obtaining a trained generator as a model for generating a 3D animation.Type: GrantFiled: November 15, 2021Date of Patent: December 5, 2023Assignee: Beijing Baidu Netcom Science Technology Co., Ltd.Inventor: Shaoxiong Yang
-
Patent number: 11830236Abstract: Provided are a method and a device for generating an avatar, an electronic equipment, a medium and a product. In the method, a to-be-detected face image of a current user is acquired. The to-be-detected face image is analyzed and at least one original component of the to-be-detected face image is obtained. Each original component of the at least one original component of the to-be-detected face image is matched with each candidate component in a component set corresponding to the each original component, and a target component corresponding to the each original component of the to-be-detected face image is obtained. The target component corresponding to the each original component of the to-be-detected face image is assembled into a personalized avatar of the current user.Type: GrantFiled: November 23, 2021Date of Patent: November 28, 2023Assignee: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.Inventor: Shaoxiong Yang
-
Patent number: 11790483Abstract: Provided are a method, an apparatus, and a device for identifying human body, including: acquiring a first original picture captured; adjusting a resolution according to the acquired picture to obtain a target picture; processing the target picture based on a preset model for human body feature point detection to determine whether the target picture includes human body information; if the target picture includes the human body information, determining human body area information in the original picture according to the human body information and inputting the human body area information into a filter, enabling that the filter determines target human body area information according to the human body area information; acquiring a next original picture captured; and determining a possible human body area in the next original picture according to the target human body area information, and performing the step of adjusting the resolution according to the possible human body area.Type: GrantFiled: September 6, 2019Date of Patent: October 17, 2023Assignee: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD.Inventors: Chen Zhao, Shaoxiong Yang, Xiaoyin Zhang
-
Patent number: 11526971Abstract: The present disclosure provides a computer-implemented method for translating an image and a computer-implemented method for training an image translation model. In the computer-implemented method for translating an image, an image translation request carrying an original image is obtained. The original image is processed to generate a pre-translated image, a mask image and a deformation parameter. The original image is deformed based on the deformation parameter to obtain a deformed image. The deformed image, the pre-translated image and the mask image are merged to generate a target translated image.Type: GrantFiled: November 30, 2020Date of Patent: December 13, 2022Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.Inventors: Shaoxiong Yang, Chen Zhao
-
Publication number: 20220375456Abstract: A method for animation synthesis includes: obtaining an audio stream to be processed and a syllable sequence, wherein both the audio stream and the syllable sequence correspond to the same text and each syllable in the syllable sequence is pinyin of each character of the text; obtaining a phoneme information sequence of the audio stream by performing phoneme detection on the audio stream, wherein each piece of phoneme information in the phoneme information sequence comprises a phoneme category and a pronunciation time period; determining a pronunciation time period corresponding to each syllable in the syllable sequence based on the syllable sequence, phoneme categories and pronunciation time periods in the phoneme information sequence; and generating an animation video corresponding to the audio stream based on the pronunciation time period corresponding to each syllable in the syllable sequence and an animation frame sequence corresponding to each syllable.Type: ApplicationFiled: June 30, 2022Publication date: November 24, 2022Inventors: Shaoxiong YANG, Chen ZHAO
-
Patent number: 11508044Abstract: A method for translating an image, a method for training an image translation model, and related electronic devices are proposed. In the method for translating an image, an image translation request carrying an original image is obtained. A down-sampled image is generated by down sampling the original image. A pre-translated image, a mask image, and deformation parameters are generated based on the down-sampled image. A size of the pre-translated image and a size of the mask image are the same as a size of the original image. A deformed image is obtained by deforming original image based on the deformation parameters. The deformed image, the pre-translated image and the mask image are fused to generate a target translation image.Type: GrantFiled: December 9, 2020Date of Patent: November 22, 2022Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.Inventors: Shaoxiong Yang, Chen Zhao
-
Patent number: 11449707Abstract: Provided are a method for processing automobile image data, an apparatus, and a readable storage medium. The method includes training a preset adversarial network according to automobile image sample data obtained by collecting and preset random noise information, to obtain a trained adversarial network for generating automobile image data; and inputting the automobile image sample data into the trained adversarial network, and generating the automobile image data including a variety of the random noise information, according to a preset generation target, wherein the automobile image data are configured to train an automobile image recognition neural network to perform automobile image recognition. Therefore, the automobile image data that may be used to train an automobile image recognition neural network are generated by using the adversarial network, which amplifies the amount of data, and reduces the cost of obtaining data.Type: GrantFiled: July 9, 2019Date of Patent: September 20, 2022Assignee: Baidu Online Network Technology (Beijing) Co., Ltd.Inventors: Shaoxiong Yang, Chen Zhao
-
Publication number: 20220180584Abstract: The present disclosure discloses a method and apparatus for generating animation. An implementation of the method may include: processing a to-be-processed material to generate a normalized text; analyzing the normalized text to generate a Chinese pinyin sequence of the normalized text; generating a reference audio based on the to-be-processed material; and obtaining a animation of facial expressions corresponding to the timing sequence of the reference audio based on the Chinese pinyin sequence and the reference audio.Type: ApplicationFiled: November 16, 2021Publication date: June 9, 2022Inventors: Shaoxiong Yang, Yang Zhao, Chen Zhao
-
Publication number: 20220084307Abstract: Provided are a method and a device for generating an avatar, an electronic equipment, a medium and a product. In the method, a to-be-detected face image of a current user is acquired. The to-be-detected face image is analyzed and at least one original component of the to-be-detected face image is obtained. Each original component of the at least one original component of the to-be-detected face image is matched with each candidate component in a component set corresponding to the each original component, and a target component corresponding to the each original component of the to-be-detected face image is obtained. The target component corresponding to the each original component of the to-be-detected face image is assembled into a personalized avatar of the current user.Type: ApplicationFiled: November 23, 2021Publication date: March 17, 2022Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.Inventor: Shaoxiong YANG
-
Publication number: 20220076470Abstract: Methods and apparatuses for generating a model and generating a 3D animation, devices, and storage mediums are provided. The method for generating a model may include: acquiring a preset sample set; acquiring pre-established generative adversarial nets, the generative adversarial nets including a generator and a discriminator; and performing training steps as follows: selecting a sample from the sample set; extracting a sample audio feature from the sample audio of the sample; inputting the sample audio feature into the generator to obtain a pseudo 3D mesh vertex sequence of the sample; inputting the pseudo 3D mesh vertex sequence and the real 3D mesh vertex sequence of the sample into the discriminator to discriminate authenticity of 3D mesh vertices; and in response to determining that the generative adversarial nets meet a training completion condition, obtaining a trained generator as a model for generating a 3D animation.Type: ApplicationFiled: November 15, 2021Publication date: March 10, 2022Inventor: Shaoxiong YANG
-
Publication number: 20210374924Abstract: The present disclosure provides a computer-implemented method for translating an image and a computer-implemented method for training an image translation model. In the computer-implemented method for translating an image, an image translation request carrying an original image is obtained. The original image is processed to generate a pre-translated image, a mask image and a deformation parameter. The original image is deformed based on the deformation parameter to obtain a deformed image. The deformed image, the pre-translated image and the mask image are merged to generate a target translated image.Type: ApplicationFiled: November 30, 2020Publication date: December 2, 2021Inventors: Shaoxiong YANG, Chen ZHAO
-
Publication number: 20210374920Abstract: A method for translating an image, a method for training an image translation model, and related electronic devices are proposed. In the method for translating an image, an image translation request carrying an original image is obtained. A down-sampled image is generated by down sampling the original image. A pre-translated image, a mask image, and deformation parameters are generated based on the down-sampled image. A size of the pre-translated image and a size of the mask image are the same as a size of the original image. A deformed image is obtained by deforming original image based on the deformation parameters. The deformed image, the pre-translated image and the mask image are fused to generate a target translation image.Type: ApplicationFiled: December 9, 2020Publication date: December 2, 2021Inventors: Shaoxiong YANG, Chen ZHAO
-
Patent number: 10983596Abstract: The present disclosure provides a gesture recognition method, a device, an electronic device, and a storage medium. The method includes: sequentially performing a recognition process on each image from a target video using a preset recognition model of palm orientation to determine a probability of containing a palm image in each image and a palm normal vector corresponding to each image; determining a group of target images from the target video based on the probability of containing the palm image in each image; and determining a target gesture corresponding to the target video based on the palm normal vector corresponding to each target image in the group of target images.Type: GrantFiled: February 14, 2020Date of Patent: April 20, 2021Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.Inventors: Chen Zhao, Shaoxiong Yang, Yuan Gao
-
Publication number: 20200301514Abstract: The present disclosure provides a gesture recognition method, a device, an electronic device, and a storage medium. The method includes: sequentially performing a recognition process on each image from a target video using a preset recognition model of palm orientation to determine a probability of containing a palm image in each image and a palm normal vector corresponding to each image; determining a group of target images from the target video based on the probability of containing the palm image in each image; and determining a target gesture corresponding to the target video based on the palm normal vector corresponding to each target image in the group of target images.Type: ApplicationFiled: February 14, 2020Publication date: September 24, 2020Inventors: Chen ZHAO, Shaoxiong YANG, Yuan GAO
-
Publication number: 20200134305Abstract: Provided are a method, an apparatus, and a device for identifying human body, including: acquiring a first original picture captured; adjusting a resolution according to the acquired picture to obtain a target picture; processing the target picture based on a preset model for human body feature point detection to determine whether the target picture includes human body information; if the target picture includes the human body information, determining human body area information in the original picture according to the human body information and inputting the human body area information into a filter, enabling that the filter determines target human body area information according to the human body area information; acquiring a next original picture captured; and determining a possible human body area in the next original picture according to the target human body area information, and performing the step of adjusting the resolution according to the possible human body area.Type: ApplicationFiled: September 6, 2019Publication date: April 30, 2020Inventors: Chen ZHAO, Shaoxiong YANG, Xiaoyin ZHANG
-
Publication number: 20190332894Abstract: Provided are a method for processing automobile image data, an apparatus, and a readable storage medium. The method includes training a preset adversarial network according to automobile image sample data obtained by collecting and preset random noise information, to obtain a trained adversarial network for generating automobile image data; and inputting the automobile image sample data into the trained adversarial network, and generating the automobile image data including a variety of the random noise information, according to a preset generation target, wherein the automobile image data are configured to train an automobile image recognition neural network to perform automobile image recognition. Therefore, the automobile image data that may be used to train an automobile image recognition neural network are generated by using the adversarial network, which amplifies the amount of data, and reduces the cost of obtaining data.Type: ApplicationFiled: July 9, 2019Publication date: October 31, 2019Applicant: Baidu Online Network Technology (Beijing) Co., Ltd.Inventors: Shaoxiong YANG, Chen ZHAO