公开(公告)号：US10861480B2

公开(公告)日：2020-12-08

申请号：US16228656

申请日：2018-12-20

Applicant: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.

Inventor： Jianwei Sun ,  Chao Li ,  Xin Li ,  Weixin Zhu ,  Ming Wen

IPC: H04R3/00 ,  H04R29/00 ,  G10L21/0272 ,  G10L21/0208 ,  G10L21/0316 ,  G10L25/78 ,  G10L25/48 ,  G10L15/06 ,  G10L21/00 ,  G10L25/21

Abstract: Embodiments of the present disclosure provide a method and a device for generating far-field speech data, a computer device and a computer readable storage medium. The method includes obtaining environmental noise in real environment and adjusting near-field speech data in a near-field speech data set based on the environmental noise, further includes generating far-field speech data based on adjusted near-field speech data and the environmental noise.

公开(公告)号：US11983849B2

公开(公告)日：2024-05-14

申请号：US17203437

申请日：2021-03-16

Applicant: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.

Inventor： Chao Li ,  Dongliang He ,  Fu Li ,  Hao Sun

IPC: G06T9/00 ,  G06F18/214 ,  G06T3/4046 ,  G06T5/00 ,  G06T11/40 ,  G06T11/60 ,  G06V10/44 ,  G06V10/774 ,  G06V10/82

CPC classification number: G06T5/005 ,  G06F18/214 ,  G06T3/4046 ,  G06T9/00 ,  G06T11/40 ,  G06T11/60 ,  G06V10/454 ,  G06V10/774 ,  G06V10/82

Abstract: An image filling method and apparatus, a device and a storage medium are disclosed. The image filling method includes: performing multilevel encoding processing on features of an image to be filled to generate multilevel encoded feature layers, sizes of the multilevel encoded feature layers being reduced layer by layer; performing layer-by-layer decoding processing on the multilevel encoded feature layers to obtain multilevel decoded feature layers and a first image, there being no missing region in the first image, wherein the layer-by-layer decoding processing includes a concatenation operation on a decoded feature layer and an encoded feature layer with a same size; and performing up-sampling processing on the first image to obtain multilevel up-sampled feature layers and a second image optimized by the up-sampling processing, the up-sampling processing including a concatenation operation on an up-sampled feature layer and a decoded feature layer with a same size.

公开(公告)号：US11227186B2

公开(公告)日：2022-01-18

申请号：US16828845

申请日：2020-03-24

Applicant: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.

Inventor： Jiadong Zhang ,  Chao Li ,  Guoyi Liu

IPC: G06K9/46 ,  G06K9/62 ,  G06T7/73

Abstract: The present disclosure relates to a method and a device for training an image recognition model and a related device. The method includes: extracting sub-image feature data from a detection frame sub-image of an input image; determining element feature data matching the sub-image feature data from an index element database; and outputting images related to the element feature data as training images for training the image recognition model. The index element database is built in advance based on a plurality of element feature data extracted from a plurality of candidate images.