公开(公告)号：US20140133550A1

公开(公告)日：2014-05-15

申请号：US14052207

申请日：2013-10-11

Applicant: STMicroelectronics S.r.l.

Inventor： Danilo Pietro Pau

IPC: H04N7/26

CPC classification number: G06K9/4671 ,  G06K9/46 ,  H04N19/124 ,  H04N19/13 ,  H04N19/132 ,  H04N19/503 ,  H04N19/93

Abstract: A first video frame and a second video frame in a flow of digital video frames are encoded by extracting for the frames in question respective sets of keypoints and descriptors, with each descriptor including a plurality of orientation histograms regarding a patch of pixels centred on the respective keypoint. Once a pair of linked descriptors has been identified, one for the first frame and one for the second frame, which have a minimum distance from among the distances between any one of the descriptors of the first frame and any one of the descriptors of the second frame, the differences of the histograms of the descriptors linked in said pair are calculated, and the descriptors linked in said pair are encoded as the set including one of the linked descriptors and the aforesaid histogram differences by subjecting the histogram differences to a thesholding setting at zero all the differences below a certain threshold, to quantization, and to an encoding of a run-length type. The run-length encoding is followed by a further encoding chosen from among a Huffman encoding, an arithmetical encoding, and a type encoding.

Abstract translation: 在数字视频帧流中的第一视频帧和第二视频帧通过提取所讨论的帧的关键点和描述符集而被编码，其中每个描述符包括关于以各自为中心的像素块的多个取向直方图 关键。 一旦已经确定了一对链接的描述符，一个对于第一帧和第二帧的一个，其具有距第一帧的任何一个描述符和第二帧的描述符中的任何一个之间的距离之间的最小距离 计算在所述对中链接的描述符的直方图的差异，并且将链接在所述对中的描述符编码为包括链接描述符之一和上述直方图差异的集合，通过对直方图差异进行保持设置 将所有差异都低于某个阈值，量化，以及运行长度类型的编码。 游程长度编码之后是从霍夫曼编码，算术编码和类型编码中选择的另外的编码。

公开(公告)号：US20130216143A1

公开(公告)日：2013-08-22

申请号：US13762267

申请日：2013-02-07

Applicant: STMicroelectronics S.r.l

Inventor： Paolo Pasteris ,  Giovanni Condello ,  Danilo Pietro Pau

IPC: G06K9/62

CPC classification number: H04N19/124 ,  G06K9/00523 ,  G06K9/00671 ,  G06K9/46 ,  G06K9/4671 ,  G06K9/6218 ,  H04N19/13 ,  H04N19/91

Abstract: One embodiment is a method for selecting and grouping key points extracted by applying a feature detector on a scene being analyzed. The method includes grouping the extracted key points into clusters that enforce a geometric relation between members of a cluster, scoring and sorting the clusters, identifying and discarding clusters that are comprised of points which represent the background noise of the image, and sub-sampling the remaining clusters to provide a smaller number of key points for the scene.

Abstract translation: 一个实施例是一种用于选择和分组通过在正在分析的场景上应用特征检测器而提取的关键点的分组的方法。 该方法包括将所提取的关键点分组成强制簇的成员之间的几何关系，对聚类进行评分和排序，识别和丢弃由表示图像的背景噪声的点组成的簇，并对 剩余的群集为场景提供较少数量的关键点。