Motion trajectory tracking for action detection

    公开(公告)号:US11488374B1

    公开(公告)日:2022-11-01

    申请号:US16578825

    申请日:2019-09-23

    Applicant: Apple Inc.

    Abstract: The disclosure pertains to techniques for image processing. One such technique comprises a method for image selection, comprising obtaining a sequence of images, detecting one or more objects in one or more images of the sequence of images, determining a location for each detected object in the one or more images, determining a trajectory for each detected object based on a determined location for each respective detected object in two or more images of the sequence of images, determining a trajectory waypoint score for the trajectory of each detected object, determining a set of selected images based on an aggregation of trajectory waypoint scores for each detected object in each respective image, and outputting the set of selected images for presentation.

    Scene motion correction in fused image systems

    公开(公告)号:US10277820B2

    公开(公告)日:2019-04-30

    申请号:US16030632

    申请日:2018-07-09

    Applicant: Apple Inc.

    Abstract: Techniques to capture and fuse short- and long-exposure images of a scene from a stabilized image capture device are disclosed. More particularly, the disclosed techniques use not only individual pixel differences between co-captured short- and long-exposure images, but also the spatial structure of occluded regions in the long-exposure images (e.g., areas of the long-exposure image(s) exhibiting blur due to scene object motion). A novel device used to represent this feature of the long-exposure image is a “spatial difference map.” Spatial difference maps may be used to identify pixels in the short-and long-exposure images for fusion and, in one embodiment, may be used to identify pixels from the short-exposure image(s) to filter post-fusion so as to reduce visual discontinuities in the output image.

    Exposure mapping and dynamic thresholding for blending of multiple images using floating exposure

    公开(公告)号:US09832378B2

    公开(公告)日:2017-11-28

    申请号:US13911243

    申请日:2013-06-06

    Applicant: Apple Inc.

    CPC classification number: H04N5/23238

    Abstract: Special blend operations for wide area-of-view image generation utilizing a “floating auto exposure” scheme are described. Pixel values in the two images being stitched together are blended within a transition band around a “seam.” identified in the overlap region between the images after changes in exposure and/or color saturation are accounted for. In some embodiments, changes in exposure and/or color saturation are accounted for through the use of one or more exposure mapping curves, the selection and use of which are based, at least in part, on a determined “Exposure Ratio” value, i.e., the amount that the camera's exposure settings have deviated from their initial capture settings. In other embodiments, the Exposure Ratio value is also used to determine regions along the seam where either: alpha blending, Poisson blending—or a combination of the two—should be used to blend in the transitional areas on each side of the seam.

    Image Registration Methods for Still Image Stabilization
    9.
    发明申请
    Image Registration Methods for Still Image Stabilization 有权
    静态图像稳定图像配准方法

    公开(公告)号:US20140363096A1

    公开(公告)日:2014-12-11

    申请号:US13911793

    申请日:2013-06-06

    Applicant: Apple Inc.

    CPC classification number: G06T7/0028 G06T7/33 G06T2207/10016 G06T2207/20016

    Abstract: Systems, methods, and computer readable media to improve image stabilization operations are described. A novel approach to pixel-based registration of non-reference images to a reference frame in a set of commonly captured images is disclosed which makes use of pyramid decomposition to more efficiently detect corners. The disclosed pixel-based registration operation may also be combined with motion sensor data-based registration approaches to register non-reference images with respect to the reference frame. When the registered non-reference images are combined with the pre-selected reference image, the resulting image is a quality stabilized image.

    Abstract translation: 描述了用于改善图像稳定操作的系统,方法和计算机可读介质。 公开了一种将非参考图像基于像素的注册到一组共同拍摄的图像中的参考帧的新颖方法,其利用金字塔分解来更有效地检测角落。 所公开的基于像素的注册操作还可以与基于运动传感器数据的注册方法组合以便相对于参考帧来注册非参考图像。 当登记的非参考图像与预先选择的参考图像组合时,所得图像是质量稳定图像。

    Unified bracketing approach for imaging

    公开(公告)号:US10977778B2

    公开(公告)日:2021-04-13

    申请号:US16426448

    申请日:2019-05-30

    Applicant: Apple Inc.

    Abstract: Devices, methods, and computer-readable media are disclosed describing an adaptive approach for image bracket selection and fusion, e.g., to generate low noise and high dynamic range (HDR) images in a wide variety of capturing conditions. An incoming image stream may be obtained from an image capture device, wherein the incoming image stream comprises a variety of differently-exposed captures, e.g., EV0 images, EV− images, EV+ images, long exposure (or synthetic long exposure) images, EV0/EV− image pairs, etc., which are received according to a particular pattern. When a capture request is received, a set of rules and/or a decision tree may be used to evaluate one or more capture conditions associated with the images from the incoming image stream and determine which two or more images to select for a fusion operation. A noise reduction process may optionally be performed on the selected images before (or after) the registration and fusion operations.

Patent Agency Ranking