专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明公开

US20240161761A1 AUDIO VISUAL SOUND SOURCE SEPARATION WITH CROSS-MODAL META CONSISTENCY LEARNING 审中-公开
公开(公告)号：US20240161761A1
公开(公告)日：2024-05-16
申请号：US18507656
申请日：2023-11-13
申请人： Md Amirul ISLAM , Seyed shahabeddin NABAVI , Irina KEZELE , Yang WANG , Yuanhao YU , Jin TANG
发明人： Md Amirul ISLAM , Seyed shahabeddin NABAVI , Irina KEZELE , Yang WANG , Yuanhao YU , Jin TANG
IPC分类号： G10L19/02 , G10L25/18 , G10L25/30 , H04N19/20
CPC分类号： G10L19/0204 , G10L25/18 , G10L25/30 , H04N19/20
摘要： System, method and/or computer readable medium for audio visual sound source Separation that applies cross-modal meta consistency learning. Inputs include an audio spectrogram that represents first sounds second sounds, a first video of a first sound producing object, and a second video of a second sound producing object. Audio features and audio tokens are generated by applying audio encoders to the audio spectrogram. Visual tokens are generated by applying a visual encoder to the first video and the second video. Respective audio-visual features are obtained based on combining the audio token with the respective visual tokens. Based on the respective audio-visual features, first and second separated audio masks are generated by applying a decoder.

2. 发明授权

US11778223B2 Method, device, and medium for generating super-resolution video 有权
公开(公告)号：US11778223B2
公开(公告)日：2023-10-03
申请号：US17406845
申请日：2021-08-19
申请人： Wentao Liu , Yuanhao Yu , Yang Wang , Juwei Lu , Xiaolin Wu , Jin Tang
发明人： Wentao Liu , Yuanhao Yu , Yang Wang , Juwei Lu , Xiaolin Wu , Jin Tang
IPC分类号： H04N19/59 , H04N19/51 , H04N19/184 , H04N19/136
CPC分类号： H04N19/51 , H04N19/136 , H04N19/184
摘要： A method, device and computer-readable medium for generating a super-resolution version of a compressed video stream. By leveraging the motion information and residual information in compressed video streams, described examples are able to skip the time-consuming motion-estimation step for most frames and make the most use of the SR results of key frames. A key frame SR module generates SR versions of I-frames and other key frames of a compressed video stream using techniques similar to existing multi-frame approaches to VSR. A non-key frame SR module generates SR version of the non-key inter frames between these key frames by making use of motion information and residual information used to encode the inter frames in the compressed video stream.

3. 发明公开

US20230169794A1 METHOD, DEVICE, AND MEDIUM FOR ADAPTIVE INFERENCE IN COMPRESSED VIDEO DOMAIN 审中-公开
公开(公告)号：US20230169794A1
公开(公告)日：2023-06-01
申请号：US17538516
申请日：2021-11-30
申请人： Irina KEZELE , Mostafa SHAHABINEJAD , Seyed shahabeddin NABAVI , Wentao LIU , Yuanhao YU , Rui Xiang CHAI , Jin TANG , Yang WANG
发明人： Irina KEZELE , Mostafa SHAHABINEJAD , Seyed shahabeddin NABAVI , Wentao LIU , Yuanhao YU , Rui Xiang CHAI , Jin TANG , Yang WANG
IPC分类号： G06V40/20 , G06V10/62 , G06N5/04
CPC分类号： G06V40/20 , G06V10/62 , G06N5/04
摘要： Methods, devices and computer-readable media for processing a compressed video to perform an inference task are disclosed. Processing the compressed video may include selecting a subset of frame encodings of the compressed video, or zero or more modalities (RGB, motion vectors, residuals) of a frame encoding, for further processing to perform the inference task. Pre-existing motion vector and/or residual information in frame encodings of the compressed video are leveraged to adaptively and efficiently perform the inference task. In some embodiments, the inference task is an action recognition task, such as a human action recognition task.

4. 发明授权

US11430138B2 Systems and methods for multi-frame video frame interpolation 有权
公开(公告)号：US11430138B2
公开(公告)日：2022-08-30
申请号：US17102114
申请日：2020-11-23
申请人： Zhixiang Chi , Rasoul Mohammadi Nasiri , Zheng Liu , Jin Tang , Juwei Lu
发明人： Zhixiang Chi , Rasoul Mohammadi Nasiri , Zheng Liu , Jin Tang , Juwei Lu
IPC分类号： G06T7/246 , G06N3/08
摘要： Systems and methods for multi-frame video frame interpolation. Higher-order motion modeling, such as cubic motion modeling, achieves predictions of intermediate optical flow between multiple interpolated frames, assisted by relaxation of the constraints imposed by the loss function used in initial optical flow estimation. A temporal pyramidal optical flow refinement module performs coarse-to-fine refinement of the optical flow maps used to generate the intermediate frames, focusing a proportionally greater amount of refinement attention to the optical flow maps for the high-error middle frames. A temporal pyramidal pixel refinement module performs coarse-to-fine refinement of the generated intermediate frames, focusing a proportionally greater amount of refinement attention to the high-error middle frames. A generative adversarial network (GAN) module calculates a loss function for training the neural networks used in the optical flow estimation module, temporal pyramidal optical flow refinement module, and/or temporal pyramidal pixel refinement module.

5. 发明申请

US20210279840A1 SYSTEMS AND METHODS FOR MULTI-FRAME VIDEO FRAME INTERPOLATION 有权
公开(公告)号：US20210279840A1
公开(公告)日：2021-09-09
申请号：US17102114
申请日：2020-11-23
申请人： Zhixiang CHI , Rasoul MOHAMMADI NASIRI , Zheng LIU , Jin TANG , Juwei LU
发明人： Zhixiang CHI , Rasoul MOHAMMADI NASIRI , Zheng LIU , Jin TANG , Juwei LU
IPC分类号： G06T3/40 , G06T7/246 , G06N3/08
摘要： Systems and methods for multi-frame video frame interpolation. Higher-order motion modeling, such as cubic motion modeling, achieves predictions of intermediate optical flow between multiple interpolated frames, assisted by relaxation of the constraints imposed by the loss function used in initial optical flow estimation. A temporal pyramidal optical flow refinement module performs coarse-to-fine refinement of the optical flow maps used to generate the intermediate frames, focusing a proportionally greater amount of refinement attention to the optical flow maps for the high-error middle frames. A temporal pyramidal pixel refinement module performs coarse-to-fine refinement of the generated intermediate frames, focusing a proportionally greater amount of refinement attention to the high-error middle frames. A generative adversarial network (GAN) module calculates a loss function for training the neural networks used in the optical flow estimation module, temporal pyramidal optical flow refinement module, and/or temporal pyramidal pixel refinement module.

6. 发明申请

US20080085046A1 Detailed identification method for articles 审中-公开
标题翻译：文章的详细识别方法
公开(公告)号：US20080085046A1
公开(公告)日：2008-04-10
申请号：US11542254
申请日：2006-10-04
申请人： Yen-Chieh Lee , Raymond R. Chen , Pen-Tsung Huang , Ming-Feng Shih , Jin-Tang Fang , Chung-Min Huang , Cheng-Chi Wu , Minh-Lun Hsieh
发明人： Yen-Chieh Lee , Raymond R. Chen , Pen-Tsung Huang , Ming-Feng Shih , Jin-Tang Fang , Chung-Min Huang , Cheng-Chi Wu , Minh-Lun Hsieh
IPC分类号： G06K9/00
CPC分类号： G07D7/128 , G06K9/00577
摘要： A detailed identification method for articles has an image-taking step and a comparison step. The image-taking step includes acts of using a portable identification device having a close-up lens and a screen to take a detailed image of an undistinguishable article and displaying the detailed image on the screen with an enlarged size. The comparison step includes acts of recalling a detailed image of a legal article from a database and displaying the detailed image on the screen of the portable identification device with an enlarged size. The two detailed images are displayed on the same screen so a user can clearly compare the detailed images to judge whether the undistinguishable article is a fake. The detailed identification method for articles has mobility and can be conveniently and extensively used.
摘要翻译：物品的详细识别方法具有摄像步骤和比较步骤。拍摄步骤包括使用具有特写镜头和屏幕的便携式识别装置拍摄不可区分的物品的详细图像并以放大的尺寸在屏幕上显示详细图像的动作。比较步骤包括从数据库调用法律文章的详细图像并以放大的尺寸在便携式识别装置的屏幕上显示详细图像的动作。两个详细图像显示在同一个屏幕上，用户可以清楚地比较详细的图像，以判断不可区分的文章是否是假的。物品的详细识别方法具有移动性并且可以方便和广泛地使用。

7. 发明申请

US20060128399A1 Method for positioning mobile station and repeater thereof 有权
标题翻译：定位移动台及其中继器的方法
公开(公告)号：US20060128399A1
公开(公告)日：2006-06-15
申请号：US10531203
申请日：2003-08-05
申请人： Weiming Duan , Jin Tang , Huiping Duan
发明人： Weiming Duan , Jin Tang , Huiping Duan
IPC分类号： H04Q7/20
CPC分类号： H04W64/00 , H04B7/155
摘要： The present invention relates to a method for positioning a mobile station and a repeater thereof. Said method comprises: the mobile communication network, when receiving a positioning request from a mobile station, instructing repeaters to send auxiliary positioning signals, and the mobile station performs measurement according to the received auxiliary positioning signals sent from the repeaters and downlink signals sent from the base station, and then estimating the position of the mobile station according to the measurement results, and thereby implementing the positioning of the mobile station. Said repeater is implemented through adding an auxiliary positioning unit in the downlink processing channel of a traditional repeater; said auxiliary positioning unit comprises a communication module, a frame timing recovery module, a timing control module, and a pilot modulating module. The present invention achieves the object of improving the accuracy of positioning the mobile station within the coverage area of repeaters; in addition, the repeater with auxiliary positioning function according to the present invention is simple to be implemented and doesn't affect the structure and signaling flow of the traditional mobile station.
摘要翻译：本发明涉及一种用于定位移动台及其中继器的方法。所述方法包括：移动通信网在接收到来自移动台的定位请求时指示中继器发送辅助定位信号，并且移动台根据从中继器发送的接收的辅助定位信号和从中继器发送的下行链路信号进行测量基站，然后根据测量结果估计移动台的位置，从而实现移动台的定位。所述中继器通过在传统中继器的下行链路处理信道中添加辅助定位单元来实现; 所述辅助定位单元包括通信模块，帧定时恢复模块，定时控制模块和导频调制模块。本发明实现了提高移动台在中继器覆盖区域内的定位精度的目的。此外，具有根据本发明的辅助定位功能的中继器简单易于实现，并且不影响传统移动台的结构和信令流。

8. 发明申请

US20050196852A1 Antibodies to calcium independent cytosolic phospholipase A2/B enzymes 失效
标题翻译：与钙无关的胞质磷脂酶A2 / B酶的抗体
公开(公告)号：US20050196852A1
公开(公告)日：2005-09-08
申请号：US10612668
申请日：2003-07-01
申请人： Simon Jones , Jin Tang
发明人： Simon Jones , Jin Tang
IPC分类号： A61K38/00 , C12N9/20 , C07H21/04
CPC分类号： C12N9/18 , A61K38/00 , C12N9/20
摘要： The invention provides a novel calcium-independent cytosolic phospholipase A2/B enzyme, polynucleotides encoding such enzyme antibodies to such enzyme, and methods for screening unknown compounds for anti-inflammatory activity mediated by the arachidonic acid cascade.
摘要翻译：本发明提供了一种新的与钙无关的胞质磷脂酶A 2 / B酶，编码这种酶的这种酶抗体的多核苷酸，以及用于筛选未知化合物用于由花生四烯酸级联介导的抗炎活性的方法。

9. 发明授权

US5589170A Calcium independent cytosolic phospholipase A.sub.2 /B enzymes 失效
公开(公告)号：US5589170A
公开(公告)日：1996-12-31
申请号：US422106
申请日：1995-04-14
申请人： Simon Jones , Jin Tang
发明人： Simon Jones , Jin Tang
IPC分类号： A61K38/00 , C12N9/20 , C12N15/55 , C12P21/04 , A61K38/46 , C07H21/04
CPC分类号： C12N9/20 , A61K38/00
摘要： The invention provides a novel calcium-independent cytosolic phospholipase A.sub.2 /B enzyme, polynucleotides encoding such enzyme and methods for screening unknown compounds for anti-inflammatory activity mediated by the arachidonic acid cascade.

10. 发明授权

US12033342B2 Methods, systems and computer medium for scene-adaptive future depth prediction in monocular videos 有权
公开(公告)号：US12033342B2
公开(公告)日：2024-07-09
申请号：US17203645
申请日：2021-03-16
申请人： Huan Liu , Zhixiang Chi , Yuanhao Yu , Yang Wang , Jin Tang
发明人： Huan Liu , Zhixiang Chi , Yuanhao Yu , Yang Wang , Jin Tang
IPC分类号： G06N3/0455 , G06F18/214 , G06N3/08 , G06T7/50 , G06T7/579 , G06V10/44
CPC分类号： G06T7/579 , G06F18/214 , G06N3/08 , G06V10/44 , G06T2207/10028 , G06T2207/20081
摘要： Systems, methods and computer-readable medium for predicting a depth for a video frame are disclosed. An example method may include steps of: receiving a plurality of training data, each comprising a set of consecutive video frames and a depth representation of a subsequent video frame to the consecutive video frames; receiving a pre-trained neural network model fθ having a plurality of weights θ; while the pre-trained neural network model fθ has not converged: computing a plurality of second weights θi′, based on each set of consecutive video frames, and updating the plurality of weights θ, based on the plurality of training data and the plurality of second weights θi′; receiving a plurality of new consecutive video frames with consecutive timestamps; and predicting a depth representation of video frame immediately subsequent to the new consecutive video frames based on the updated plurality of weights θ.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式