-
公开(公告)号：EP4391560A1
公开(公告)日：2024-06-26
申请号：EP22925684.7
申请日：2022-11-15
发明人： ZHAO, Yuanyuan , REN, Jingjing , LI, Chen
IPC分类号： H04N21/44
CPC分类号： H04N19/132 , H04N19/167 , H04N19/17 , H04N19/124 , G06V20/40 , G06V10/26 , G06V10/7715
摘要： The present application relates to the technical field of video processing. Disclosed are a video processing method and apparatus, a device and a storage medium. The method comprises: inputting a video to be processed and a video mask into a coding model for feature extraction to obtain a first video feature vector and a second video feature vector (202);determining an index distribution of a first quantized feature vector in a discrete latent space formed on the basis of the first quantized feature vector (204); determining a second quantized feature vector in the discrete latent space on the basis of the second video feature vector and the index distribution(206);and inputting the first quantized feature vector and the second quantized feature vector into a decoding model to obtain a reconstructed video (208).
-
2.发明公开
公开(公告)号：EP3920146A1
公开(公告)日：2021-12-08
申请号：EP20748016.1
申请日：2020-01-14
发明人： SUN, Shuang , LI, Chen , TAI, Yuwing , JIA, Jiaya , SHEN, Xiaoyong
IPC分类号： G06T13/40
摘要： A method for training an SMPL parameter prediction model, comprising: obtaining a sample picture; inputting the sample picture to a posture parameter prediction model to obtain a posture prediction parameter; inputting the sample picture to a morphology parameter prediction model to obtain a morphology prediction parameter; computing a model prediction loss according to an SMPL prediction parameter and annotation information of the sample picture; and inversely training the posture parameter prediction model and the morphology parameter prediction model according to the model prediction loss.
-
公开(公告)号：EP3745358A1
公开(公告)日：2020-12-02
申请号：EP19853856.3
申请日：2019-06-24
发明人： LI, Chen , YI, Hongwei , CAO, Qiong , SHEN, Xiaoyong , TAI, Yuwing , JIA, Jiaya
IPC分类号： G06T17/00
摘要： Embodiments of this application disclose a method and an apparatus for reconstructing a three-dimensional model of a human body, and a storage medium. According to the embodiments of this application, a two-dimensional image of a human body to be reconstructed may be obtained; the two-dimensional image is mapped to a voxel space, to obtain a voxelized result of the two-dimensional image; then calculation is performed on the voxelized result by using a preset three-dimensional parameter regression network, to obtain three-dimensional morphable model parameters corresponding to the two-dimensional image; and a three-dimensional model of the human body is reconstructed based on the three-dimensional morphable model parameters of the two-dimensional image.
-
4.发明公开
公开(公告)号：EP3783571A1
公开(公告)日：2021-02-24
申请号：EP19788578.3
申请日：2019-03-06
发明人： PENG, Weilong , LI, Chen , TAI, Yu Wing , SHEN, Xiaoyong , JIA, Jiaya
IPC分类号： G06T17/00
摘要： Disclosed by the embodiments of the present invention are a face model processing method and device, a nonvolatile computer-readable storage medium and an electronic device, the method comprising the following steps: acquiring a three-dimensional face model corresponding to a user picture, and acquiring a selected sample oral cavity model from an oral cavity model library; registering the sample oral cavity model into the three-dimensional face model using an oral cavity position parameter in the three-dimensional face model; adjusting the oral cavity morphology of the registered sample oral cavity model by using an expression parameter of the three-dimensional face model so as to generate a target oral cavity model; and generating a three-dimensional facial expression model corresponding to the user picture on the basis of the three-dimensional face model and the target oral cavity model.
-
-
-
搜索结果
4 条记录 (耗时 0.015 秒)
