-
公开(公告)号：WO2011046710A1
公开(公告)日：2011-04-21
申请号：PCT/US2010/049350
申请日：2010-09-17
Applicant: APPLE INC. , WU, Hsi-Jung , CHUNG, Chris, Yoochang , SHI, Xiaojin , NORMILE, James
Inventor： WU, Hsi-Jung , CHUNG, Chris, Yoochang , SHI, Xiaojin , NORMILE, James
CPC classification number: G06K9/00228 , G06T3/00 , G06T5/006 , G06T2207/20104 , G06T2207/30201 , H04N5/217 , H04N5/23219 , H04N5/23258 , H04N5/23267
Abstract: Methods, systems, and apparatus are presented for reducing distortion in an image, such as a video image. A video image can be captured by an image capture device, e.g. during a video conferencing session. Distortion correction processing, such as the application of one or more warping techniques, can be applied to the captured image to produce a distortion corrected image, which can be transmitted to one or more participants. The warping techniques can be performed in accordance with one or more warp parameters specifying a transformation of the captured image. Further, the warp parameters can be generated in accordance with an orientation of the image capture device, which can be determined based on sensor data or can be a fixed value. Additionally or alternatively, the warp parameters can be determined in accordance with a reference image or model to which the captured image should be warped.
Abstract translation: 呈现了用于减少诸如视频图像的图像中的失真的方法，系统和装置。 视频图像可以由图像捕获设备捕获，例如， 在视频会议期间。 畸变校正处理，例如应用一个或多个翘曲技术，可以应用于所捕获的图像，以产生可以发送到一个或多个参与者的失真校正图像。 翘曲技术可以根据指定捕获图像的变换的一个或多个翘曲参数来执行。 此外，可以根据可以基于传感器数据确定的图像捕获装置的取向来生成翘曲参数，或者可以是固定值。 附加地或替代地，可以根据捕获的图像应该翘曲的参考图像或模型来确定翘曲参数。
-
2.发明申请
公开(公告)号：WO2013184487A2
公开(公告)日：2013-12-12
申请号：PCT/US2013/043395
申请日：2013-05-30
Applicant: APPLE, INC.
Inventor： LEONTARIS, Athanasios , GUO, Haitao , SHI, Xiaojin
IPC: H04N7/26
CPC classification number: H04N21/44209 , H04N19/107 , H04N19/164 , H04N19/172 , H04N19/196 , H04N19/65 , H04N19/89 , H04N21/6379
Abstract: An error recovery method may be engaged by an encoder to recover from misalignment between reference picture caches at the encoder and decoder. When a communication error is detected between a coder and a decoder, a number of non-acknowledged reference frames present in the decoder's reference picture cache may be estimated. Thereafter, frames may be coded as reference frames in a number greater or equal to the number of non-acknowledged reference frames that are estimated to be present in the decoder's reference picture cache. Thereafter, ordinary coding operations may resume. Typically, a final reference frame that is coded in the error recovery mode will be coded as a synchronization frame that has high coding quality. The coded reference frames that precede it may be coded at low quality (or may be coded as SKIP-coded frames). On reception and decoding, the preceding frames may cause the decoder to flush from its reference picture cache any non-acknowledged reference frames that otherwise might collide with the new synchronization frame. In this manner, alignment between the encoder and decoder may be restored.
Abstract translation: 编码器可以接合错误恢复方法以从编码器和解码器处的参考图像高速缓存之间的未对准中恢复。 当在编码器和解码器之间检测到通信错误时，可以估计存在于解码器的参考图像高速缓存中的多个未确认参考帧。 此后，帧可以被编码为大于或等于估计存在于解码器的参考图像缓存中的未确认参考帧的数量的参考帧。 此后，可以恢复普通编码操作。 通常，以错误恢复模式编码的最终参考帧将被编码为具有高编码质量的同步帧。 其之前的编码参考帧可以以低质量编码（或者可以被编码为SKIP编码的帧）。 在接收和解码时，前述帧可能导致解码器从其参考图像缓存中刷新否则可能与新的同步帧相冲突的任何未确认的参考帧。 以这种方式，可以恢复编码器和解码器之间的对准。
-
公开(公告)号：WO2008049052A2
公开(公告)日：2008-04-24
申请号：PCT/US2007/081758
申请日：2007-10-18
Applicant: APPLE INC. , HASKELL, Barin Geoffry , WU, Hsi-jung , SHI, Xiaojin
Inventor： WU, Hsi-jung , SHI, Xiaojin
Abstract: A first improvement is described for prediction of motion vectors to be used in prediction of video data for enhancement layer data. Arbitrary pixelblock partitioning between base layer data and enhancement layer data raises problems to identify base layer motion vectors to be used as prediction sources for enhancement layer motion vectors. The disclosed method develops enhancement layer motion vectors by scaling a base layer pixelblock partition map according to a size difference between the base layer video image and the enhancement layer video image, then identified scale base layer pixelblocks that are co-located with the enhancement layer pixelblocks for which motion vector prediction is to be performed. Motion vectors from the scaled co-located base layer pixelblocks are averaged, weighted according to a degree of overlap between the base layer pixelblocks and the enhancement layer pixelblock. Another improvement is obtained by filtering recovered base layer image data before being provided to an enhancement layer decoder. When a specified filter requires image data outside a prediction region available from a base layer decoder, the prediction region data may be supplemented with previously-decoded data from an enhancement layer at a border of the prediction region.
Abstract translation: 描述了用于预测用于增强层数据的视频数据的预测中的运动矢量的第一改进。 基层数据和增强层数据之间的任意像素块划分提出了一些问题，用于识别要用作增强层运动矢量的预测源的基本层运动矢量。 所公开的方法通过根据基本层视频图像和增强层视频图像之间的尺寸差来缩放基本层像素块分割图来开发增强层运动矢量，然后识别与增强层像素块共同定位的比例基底层像素块 进行运动矢量预测。 根据基层像素块和增强层像素块之间的重叠程度对来自缩放的同位置的基本层像素块的运动矢量进行平均化。 通过在提供给增强层解码器之前对恢复的基层图像数据进行滤波来获得另一改进。 当指定的滤波器需要从基本层解码器可用的预测区域之外的图像数据时，可以从预测区域的边界处的增强层补充预先解码的数据。
-
公开(公告)号：WO2020236904A1
公开(公告)日：2020-11-26
申请号：PCT/US2020/033750
申请日：2020-05-20
Applicant: APPLE INC.
Inventor： ROSSI, Francesco , JAGADEESH, Vignesh , SHARMA, Vinay , ZULIANI, Marco , SHI, Xiaojin , POULAIN, Benjamin
Abstract: A method includes receiving input data at a trained machine learning model that includes a common part and task-specific parts, receiving an execution instruction that identifies one or more processing tasks to be performed, processing the input data using the common part of the trained machine learning model to generate intermediate data, and processing the intermediate data using one or more of the task-specific parts of the trained machine learning model based on the execution instruction to generate one or more outputs.
-
公开(公告)号：WO2009149151A2
公开(公告)日：2009-12-10
申请号：PCT/US2009/046079
申请日：2009-06-03
Applicant: APPLE INC. , SHI, Xiaojin , WU, Hsi-jung
Inventor： SHI, Xiaojin , WU, Hsi-jung
CPC classification number: H04N19/86 , H04N19/103 , H04N19/127 , H04N19/154 , H04N19/156 , H04N19/164 , H04N19/174 , H04N19/176 , H04N19/196 , H04N19/436 , H04N19/46 , H04N19/463 , H04N19/61
Abstract: Embodiments of the present invention provide apparatuses and methods of coding video. The apparatuses and methods may further provide coding a source video sequence according to a block-based coding process, estimating processing capabilities of a target decoder, determining if the estimated processing capabilities are sufficient to perform deblocking filtering. If not sufficient, the apparatuses and methods may provide computing deblocking filter strengths for pixel blocks of the source video sequence to be used at decoding, and transmitting the deblocking filter strengths in a coded video data signal with the coded video data. Moreover, if not sufficient, the apparatuses and methods may provide changing coding parameters including, but not limited to, block sizes, transform sizes, and Qmatrix.
Abstract translation: 本发明的实施例提供了对视频进行编码的装置和方法。 装置和方法可以进一步提供根据基于块的编码处理对源视频序列进行编码，估计目标解码器的处理能力，确定所估计的处理能力是否足以执行去块滤波。 如果不足够，则设备和方法可以为解码时要使用的源视频序列的像素块提供计算解块滤波器强度，并将编码的视频数据信号中的解块滤波器强度传送到编码的视频数据。 此外，如果不足够，则设备和方法可以提供改变的编码参数，包括但不限于块大小，变换大小和Qmatrix。
-
Patent Agency Ranking
公开(公告)号：WO2020242686A1
公开(公告)日：2020-12-03
申请号：PCT/US2020/029937
申请日：2020-04-24
Applicant: APPLE INC.
Inventor： PAEK, Timothy S. , ROSSI, Francesco , DHANANI, Jamil , AVERY, Keith P. , JEONG, Minwoo , SHI, Xiaojin , KAUR, Harveen , WESTING, Brandt M.
Abstract: The subject technology receives a neural network model in a model format, the model format including information for a set of layers of the neural network model, each layer of the set of layers including a set of respective operations. The subject technology generates neural network (NN) code from the neural network model, the NN code being in a programming language distinct from the model format, and the NN code comprising a respective memory allocation for each respective layer of the set of layers of the neural network model, where the generating comprises determining the respective memory allocation for each respective layer based at least in part on a resource constraint of a target device. The subject technology compiles the NN code into a binary format. The subject technology generates a package for deploying the compiled NN code on the target device.
-
公开(公告)号：WO2008033830A3
公开(公告)日：2008-03-20
申请号：PCT/US2007/078148
申请日：2007-09-11
Applicant: APPLE INC. , NORMILE, Jim , PUN, Thomas , SHI, Xiaojin , TONG, Xin , WU, Hsi-jung
Inventor： NORMILE, Jim , PUN, Thomas , SHI, Xiaojin , TONG, Xin , WU, Hsi-jung
IPC: H04N7/26
Abstract: Techniques for encoding data based at least in part upon an awareness of the decoding complexity of the encoded data and the ability of a target decoder to decode the encoded data are disclosed. In some embodiments, a set of data is encoded based at least in part upon a state of a target decoder to which the encoded set of data is to be provided. In some embodiments, a set of data is encoded based at least in part upon the states of multiple decoders to which the encoded set of data is to be provided.
-
公开(公告)号：WO2016025713A1
公开(公告)日：2016-02-18
申请号：PCT/US2015/045055
申请日：2015-08-13
Applicant: APPLE INC.
Inventor： TANG, Feng , LI, Ang , SHI, Xiaojin
CPC classification number: G06K9/00355 , G06F3/017 , G06F3/0304 , G06F3/0425 , G06K9/4609 , G06T7/246 , G06T7/254 , G06T2200/04 , G06T2207/30196 , H04N13/207 , H04N13/271 , H04N2013/0085
Abstract: In the field of Human-computer interaction (HCI), i.e., the study of the interfaces between people (i.e., users) and computers, understanding the intentions and desires of how the user wishes to interact with the computer is a very important problem. The ability to understand human gestures, and, in particular, hand gestures, as they relate to HCI, is a very important aspect in understanding the intentions and desires of the user in a wide variety of applications. In this disclosure, a novel system and method for three-dimensional hand tracking using depth sequences is described. Some of the major contributions of the hand tracking system described herein include: 1.) a robust hand detector that is invariant to scene background changes; 2.) a bi-directional tracking algorithm that prevents detected hands from always drifting closer to the front of the scene (i.e., forward along the z-axis of the scene); and 3.) various hand verification heuristics.
Abstract translation: 在人机交互（HCI）领域，即研究人（即用户）和计算机之间的接口，理解用户希望如何与计算机交互的意图和期望是非常重要的问题。 了解人类手势，特别是手势，与HCI相关的能力在了解用户在各种应用中的意图和欲望方面是非常重要的一个方面。 在本公开中，描述了使用深度序列的三维手跟踪的新颖系统和方法。 本文描述的手动跟踪系统的一些主要贡献包括：1.）对场景背景变化不变的鲁棒手指检测器; 2.）双向跟踪算法，其防止检测到的手总是漂移到靠近场景的前方（即，沿着场景的z轴向前）; 和3.）各种手验证启发式。
-
9.发明申请
公开(公告)号：WO2010040013A1
公开(公告)日：2010-04-08
申请号：PCT/US2009/059307
申请日：2009-10-02
Applicant: APPLE INC. , HASKELL, Barin , SHI, Xiaojin
Inventor： HASKELL, Barin , SHI, Xiaojin
IPC: H04N7/26
CPC classification number: H04N19/156 , H04N19/103 , H04N19/12 , H04N19/132 , H04N19/147 , H04N19/15 , H04N19/154 , H04N19/176
Abstract: Systems and methods for applying a new quality metric for coding video are provided. The metric, based on the Just Noticeable Difference (JND) distortion visibility model, allows for efficient selection of coding techniques that limit perceptible distortion in the video while still taking into account parameters, such as desired bit rate, that can enhance system performance. Additionally, the unique aspects of each input type, system and display may be considered. Allowing for a programmable minimum viewing distance (MVD) parameter also ensures that the perceptible distortion will not be noticeable at the specified MVD, even though the perceptible distortion may be significant at an alternate distance.
Abstract translation: 提供了应用用于编码视频的新质量度量的系统和方法。 基于JustStable Difference（JND）失真可见性模型的度量能够有效地选择限制视频中可察觉的失真的编码技术，同时仍然考虑可以提高系统性能的所需比特率等参数。 此外，可以考虑每个输入类型，系统和显示器的独特方面。 允许可编程的最小观察距离（MVD）参数还确保在指定的MVD下可感知的失真不会显着，即使可察觉的失真在另一距离处可能是显着的。
-
公开(公告)号：WO2008092076A2
公开(公告)日：2008-07-31
申请号：PCT/US2008/052044
申请日：2008-01-25
Applicant: APPLE INC. , SHI, Xiaojin , WU, Hsi-Jung , NORMILE, Jim
Inventor： SHI, Xiaojin , WU, Hsi-Jung , NORMILE, Jim
IPC: H04N7/26
CPC classification number: H04N21/23439 , H04N19/156 , H04N19/30 , H04N19/40 , H04N19/46 , H04N21/234327 , H04N21/235 , H04N21/2662 , H04N21/435 , H04N21/84
Abstract: Systems, apparatuses and methods whereby coded bitstreams are delivered to downstream end-user devices having various performance capabilities. A head-end encoder/video store generates a primary coded bitstream and metadata for delivery to an intermediate re-encoding system. The re-encoding system recodes the primary coded bitstream to generate secondary coded bitstreams based on coding parameters in the metadata. Each secondary coded bitstream is matched to a conformance point of a downstream end-user device. Coding parameters for each conformance point can be derived from the head-end encoder encoding original source video to generate the secondary coded bitstreams and extracting information from the coding process/results. The metadata can then can be communicated as part of the primary coded bitstream (e.g., as SEI) or can be communicated separately. As a result, the complexity of the secondary coded bitstream is appropriately scaled to match the capabilities of the downstream end-user device to which it is delivered.
Abstract translation: 编码比特流由此被传送到具有各种性能能力的下游终端用户设备的系统，设备和方法。 头端编码器/视频存储器生成主编码比特流和元数据以传送到中间重编码系统。 重新编码系统重新编码主要编码比特流以基于元数据中的编码参数来生成次要编码比特流。 每个二级编码比特流与下游终端用户设备的一致性点匹配。 可以从编码原始源视频的头端编码器导出用于每个一致性点的编码参数，以生成二次编码的比特流并从编码过程/结果中提取信息。 元数据然后可以作为主要编码比特流的一部分（例如，作为SEI）被传送，或者可以被分开传送。 结果，二级编码比特流的复杂度被适当地缩放以匹配其被递送到的下游最终用户设备的能力。
-
-
-
-
-
-
-
-
-
Search Results
17
records
(took 0.024 seconds)
