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. 发明授权

US10034005B2 Banding prediction for video encoding 有权
公开(公告)号：US10034005B2
公开(公告)日：2018-07-24
申请号：US14732551
申请日：2015-06-05
申请人： Xun Xu , Peng Ye
发明人： Xun Xu , Peng Ye
IPC分类号： H04N19/176 , H04N19/103 , H04N19/117 , H04N19/154
摘要： A system and method for facilitating inhibiting banding in video data, in part by anticipating when banding may occur given certain encoder parameters. An example method includes receiving an input stream of video data; extracting feature information characterizing the stream of video data; using a feed-forward neural network to process the feature information to estimate when a particular block of the stream of video data will exhibit an artifact when encoded using certain Quantization Parameters (QPs); incorporating the indication into metadata associated with the stream of video data; and transferring the stream of video data and metadata to a video encoder. In a more specific embodiment, the feature information includes color information and texture information, and the neural network includes a feed-forward neural network that includes a classifier with a sigmoid activation function, and which has been trained using a cross-entropy cost function.

2. 发明申请

US20160360202A1 BANDING PREDICTION FOR VIDEO ENCODING 审中-公开
标题翻译：用于视频编码的带宽预测
公开(公告)号：US20160360202A1
公开(公告)日：2016-12-08
申请号：US14732551
申请日：2015-06-05
申请人： Xun Xu , Peng Ye
发明人： Xun Xu , Peng Ye
IPC分类号： H04N19/126 , H04N19/89 , H04N19/196 , H04N19/98 , H04N19/159 , H04N19/187
CPC分类号： H04N19/176 , H04N19/103 , H04N19/117 , H04N19/154
摘要： A system and method for facilitating inhibiting banding in video data, in part by anticipating when banding may occur given certain encoder parameters. An example method includes receiving an input stream of video data; extracting feature information characterizing the stream of video data; using a feed-forward neural network to process the feature information to estimate when a particular block of the stream of video data will exhibit an artifact when encoded using certain Quantization Parameters (QPs); incorporating the indication into metadata associated with the stream of video data; and transferring the stream of video data and metadata to a video encoder. In a more specific embodiment, the feature information includes color information and texture information, and the neural network includes a feed-forward neural network that includes a classifier with a sigmoid activation function, and which has been trained using a cross-entropy cost function.
摘要翻译：一种有助于抑制视频数据中的条带化的系统和方法，部分是通过预期在给定某些编码器参数的情况下可能发生条带的发生。一种示例性方法包括：接收视频数据的输入流; 提取表征视频数据流的特征信息; 使用前馈神经网络处理特征信息以估计当使用某些量化参数（QP）进行编码时，视频数据流的特定块何时将呈现伪影; 将该指示结合到与视频数据流相关联的元数据中; 并将视频数据和元数据流传送到视频编码器。在更具体的实施例中，特征信息包括颜色信息和纹理信息，并且神经网络包括前馈神经网络，其包括具有S形激活功能的分类器，并且已经使用交叉熵代价函数训练。

3. 发明授权

US09741107B2 Full reference image quality assessment based on convolutional neural network 有权
公开(公告)号：US09741107B2
公开(公告)日：2017-08-22
申请号：US14732518
申请日：2015-06-05
申请人： Xun Xu , Peng Ye
发明人： Xun Xu , Peng Ye
IPC分类号： G06T7/00 , G06T5/00 , G06K9/66 , G06K9/46 , G06K9/62 , G06N3/08
CPC分类号： G06T7/0002 , G06K9/036 , G06K9/46 , G06K9/4628 , G06K9/6215 , G06K9/6272 , G06K9/66 , G06N3/08 , G06T5/002 , G06T2207/10016 , G06T2207/20081 , G06T2207/20084 , G06T2207/30168
摘要： Embodiments generally relate to providing systems and methods for assessing image quality of a distorted image relative to a reference image. In one embodiment, the system comprises a convolutional neural network that accepts as an input the distorted image and the reference image, and provides as an output a metric of image quality. In another embodiment, the method comprises inputting the distorted image and the reference image to a convolutional neural network configured to process the distorted image and the reference image and provide as an output a metric of image quality.

4. 发明申请

US20160358321A1 FULL REFERENCE IMAGE QUALITY ASSESSMENT BASED ON CONVOLUTIONAL NEURAL NETWORK 有权
标题翻译：基于互联神经网络的完整参考图像质量评估
公开(公告)号：US20160358321A1
公开(公告)日：2016-12-08
申请号：US14732518
申请日：2015-06-05
申请人： Xun Xu , Peng Ye
发明人： Xun Xu , Peng Ye
IPC分类号： G06T7/00 , G06N3/08 , G06K9/46 , G06K9/62 , G06T5/00 , G06K9/66
CPC分类号： G06T7/0002 , G06K9/036 , G06K9/46 , G06K9/4628 , G06K9/6215 , G06K9/6272 , G06K9/66 , G06N3/08 , G06T5/002 , G06T2207/10016 , G06T2207/20081 , G06T2207/20084 , G06T2207/30168
摘要： Embodiments generally relate to providing systems and methods for assessing image quality of a distorted image relative to a reference image. In one embodiment, the system comprises a convolutional neural network that accepts as an input the distorted image and the reference image, and provides as an output a metric of image quality. In another embodiment, the method comprises inputting the distorted image and the reference image to a convolutional neural network configured to process the distorted image and the reference image and provide as an output a metric of image quality.
摘要翻译：实施例通常涉及提供用于评估失真图像相对于参考图像的图像质量的系统和方法。在一个实施例中，系统包括卷积神经网络，其接受失真图像和参考图像作为输入，并且提供图像质量的度量作为输出。在另一个实施例中，该方法包括将畸变图像和参考图像输入到被配置为处理失真图像和参考图像的卷积神经网络，并提供图像质量的度量作为输出。

5. 发明申请

US20140136121A1 METHOD FOR ASSEMBLING SEQUENCED SEGMENTS 审中-公开
标题翻译：组装序列部分的方法
公开(公告)号：US20140136121A1
公开(公告)日：2014-05-15
申请号：US14130706
申请日：2011-07-05
申请人： Xun Xu , Ye Tao , Zequn Zheng , Jun Wang
发明人： Xun Xu , Ye Tao , Zequn Zheng , Jun Wang
IPC分类号： G06F19/22
CPC分类号： G16B30/00 , G16B40/00
摘要： The present invention relates to a method for optimizing the assembled result of sequencing data using a genetic map. In particular, provided in the present invention is a new method for assembling individual sequenced segments, which comprises the step of constructing the genetic map with a genetic marker. Furthermore, also provided in the present invention is a method for assembling the individual sequenced segments into a genome sequence, such as a chromosome sequence.
摘要翻译：本发明涉及使用遗传图谱优化测序数据的组合结果的方法。特别地，本发明提供了一种用于组装单个测序片段的新方法，其包括用遗传标记构建遗传图谱的步骤。此外，本发明还提供了将各个测序片段组装成基因组序列（例如染色体序列）的方法。

6. 发明申请

US20130243308A1 INTEGRATED INTERACTIVE SEGMENTATION WITH SPATIAL CONSTRAINT FOR DIGITAL IMAGE ANALYSIS 有权
标题翻译：用于数字图像分析的空间约束的集成交互分段
公开(公告)号：US20130243308A1
公开(公告)日：2013-09-19
申请号：US13423209
申请日：2012-03-17
申请人： Su Wang , Shengyang Dai , Xun Xu , Akira Nakamura , Takeshi Ohashi , Jun Yokono
发明人： Su Wang , Shengyang Dai , Xun Xu , Akira Nakamura , Takeshi Ohashi , Jun Yokono
IPC分类号： G06K9/62
CPC分类号： G06T7/0081 , G06K9/00 , G06T7/11 , G06T7/143 , G06T7/162 , G06T7/194 , G06T2207/20081
摘要： An integrated interactive segmentation with spatial constraint method utilizes a combination of several of the most popular online learning algorithms into one and implements a spatial constraint which defines a valid mask local to the user's given marks. Additionally, both supervised learning and statistical analysis are integrated, which are able to compensate each other. Once prediction and activation are obtained, pixel-wised multiplication is conducted to fully indicate how likely each pixel belongs to the foreground or background.
摘要翻译：具有空间约束方法的集成交互式分割将几种最流行的在线学习算法的组合应用于一种，并实现了定义用户给定标记本地的有效掩码的空间约束。此外，监督学习和统计分析都是综合的，可以相互补偿。一旦获得了预测和激活，就进行像素增益乘法以完全指示每个像素属于前景或背景的可能性。

7. 发明申请

US20130243277A1 FLOURESCENT DOT COUNTING IN DIGITAL PATHOLOGY IMAGES 有权
标题翻译：数字病理图像中的食物点数
公开(公告)号：US20130243277A1
公开(公告)日：2013-09-19
申请号：US13423208
申请日：2012-03-17
申请人： Su Wang , Xun Xu , Akira Nakamura
发明人： Su Wang , Xun Xu , Akira Nakamura
IPC分类号： G06K9/00
CPC分类号： G06K9/0014 , G06K9/00147
摘要： Fluorescence in situ hybridization (FISH) enables the detection of specific DNA sequences in cell chromosomes by the use of selective staining. Due to the high sensitivity, FISH allows the use of multiple colors to detect multiple targets simultaneously. The target signals are represented as colored dots, and enumeration of these signals is called dot counting. Using a two-stage segmentation framework guarantees locating all potential dots including overlapped dots.
摘要翻译：荧光原位杂交（FISH）能够通过选择性染色检测细胞染色体中的特定DNA序列。由于高灵敏度，FISH允许使用多种颜色同时检测多个目标。目标信号表示为彩色点，这些信号的枚举称为点计数。使用两阶段分割框架可确保定位所有可能的点，包括重叠点。

8. 发明申请

US20130060713A1 Receipt digitizing method for retail customers 审中-公开
标题翻译：零售客户的收据数字化方法
公开(公告)号：US20130060713A1
公开(公告)日：2013-03-07
申请号：US13507291
申请日：2012-06-18
申请人： Xun Xu
发明人： Xun Xu
IPC分类号： G06Q40/00
CPC分类号： G06Q30/02 , G06Q40/02
摘要： Method of receipt digitizing is disclosed. Retail customers create digital copies of paper receipts with a kiosk system before they leave. The digital copies are sent to customers' online space. This low cost method benefits customers with a centralized and organized means of keeping their receipts, and provides retailers a better way of delivering promotion information to relevant recipients. The solution is also good to the environment because it cuts coupon printing and increases awareness of the convenience of digital receipts. This awareness will eventually lead to the final phase out of paper receipt printing.
摘要翻译：公开了收据数字化的方法。零售客户在离开之前用信息亭系统创建纸质数字副本。数字副本将发送给客户的在线空间。这种低成本方法使客户能够以集中和有组织的方式保留收据，并为零售商提供将相关收件人的促销信息提供给更好的方式。该解决方案对环境也很好，因为它减少了优惠券打印，并提高了数字收据方便性的意识。这种意识将最终导致纸张收据印刷的最后阶段。

9. 发明授权

US07200282B2 Implementation of Hough transform and its application in line detection and video motion analysis 有权
标题翻译：霍夫变换的实现及其在线检测和视频运动分析中的应用
公开(公告)号：US07200282B2
公开(公告)日：2007-04-03
申请号：US10319767
申请日：2002-12-12
申请人： Xun Xu
发明人： Xun Xu
IPC分类号： G06K9/36 , G06K9/46
CPC分类号： G06K9/4633 , G06T7/215 , G06T7/262
摘要： An implementation of Hough transform is described. According to one aspect, the Hough transform is used for line detection in an image based on a one-dimensional voting array.
摘要翻译：描述了霍夫变换的实现。根据一个方面，霍夫变换用于基于一维投票阵列的图像中的线检测。

10. 发明授权

US07136509B2 Implementation of Hough transform and its application in video motion analysis 有权
标题翻译：霍夫变换的实现及其在视频运动分析中的应用
公开(公告)号：US07136509B2
公开(公告)日：2006-11-14
申请号：US10318857
申请日：2002-12-12
申请人： Xun Xu
发明人： Xun Xu
IPC分类号： G06K9/00
CPC分类号： G06K9/4633 , G06T7/20 , G06T7/262
摘要： An implementation of Hough transform is described. According to one aspect, the Hough transform is used to perform motion analysis of patterns in a video using one-dimensional detectors and a one-dimensional voting array.
摘要翻译：描述了霍夫变换的实现。根据一个方面，霍夫变换用于使用一维检测器和一维投票阵列对视频中的图案进行运动分析。

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