发明名称 Detecting objects in a sequence of images
摘要 The subject matter of this specification can be implemented in, among other things, a computer-implemented method for object detection. The method includes detecting at least one initial position of an object of a specific type in at least one initial image within an ordered sequence of images stored in a memory. The method further includes detecting at least one subsequent position of the object in at least one subsequent image within the ordered sequence of images. The method further includes estimating one or more intermediate positions of the object in one or more intermediate images within the ordered sequence of images between the initial image and the subsequent image based on the initial position and the subsequent position.
申请公布号 US8977003(B1) 申请公布日期 2015.03.10
申请号 US201213551469 申请日期 2012.07.17
申请人 Google Inc. 发明人 Kwan Gavan;Doshi Rushabh Ashok;Lebrón Federico Javier
分类号 G06K9/00 主分类号 G06K9/00
代理机构 Lowenstein Sandler LLP 代理人 Lowenstein Sandler LLP
主权项 1. A method comprising: detecting positions of instances of an object of a specific type within an ordered sequence of images stored in a memory, wherein the ordered sequence of images comprises a video, and wherein the positions of the instances of the object comprise at least one initial position of at least one initial instance of the object in at least one initial image within the ordered sequence of images and at least one subsequent position of at least one subsequent instance of the object in at least one subsequent image within the ordered sequence of images after the initial image in the ordered sequence of images; determining that the ordered sequence of images comprises a number of intermediate images between the initial image and the subsequent image that is within a first threshold number of images, wherein the first threshold number of images is based on a first threshold amount of time; estimating, by one or more processing devices, one or more intermediate positions of one or more intermediate instances of the object in the intermediate images based on the initial position and the subsequent position in response to determining that the number of intermediate images is within the first threshold number of images; buffering the video for at least the first threshold amount of time; performing an operation on the initial instance, the intermediate instances, and the subsequent instance of the object in real-time with a delay due to buffering the video for at least the first threshold amount of time; and providing the video for presentation.
地址 Mountain View CA US