Gopro, Inc.

Timothy MacMillan

Scott Patrick Campbell

David A Newman

Balineedu Chowdary Adsumilli

Yajie Sun

Adeel Abbas

Sumit Chawla

Zhinian Jing

Cesar Douady

Elizabeth Torres Edwards

Elizabeth Torres Martin

Apparatus and methods for stitching images, or re-stitching previously stitched images. Specifically, the disclosed systems in one implementation save stitching information and/or original overlap source data during an original stitching process. During subsequent retrieval, rendering, and/or display of the stitched images, the originally stitched image can be flexibly augmented, and/or re-stitched to improve the original stitch quality. Practical applications of the disclosed solutions enable, among other things, a user to create and stitch a wide field of view (FOV) panorama from multiple source images on a device with limited processing capability (such as a mobile phone or other capture device). Moreover, post-processing stitching allows for the user to convert from one image projection to another without fidelity loss (or with an acceptable level of loss).

Apparatus and methods for the storage of overlapping regions of imaging data for the generation of optimized stitched images

A system captures a first hemispherical image and a second hemispherical image, each hemispherical image including an overlap portion, the overlap portions capturing a same field of view, the two hemispherical images collectively comprising a spherical FOV and separated along a longitudinal plane. The system maps a modified first hemispherical image to a first portion of the 2D projection of a cubic image, the modified first hemispherical image including a non-overlap portion of the first hemispherical image, and maps a modified second hemispherical image to a second portion of the 2D projection of the cubic image, the modified second hemispherical image also including a non-overlap portion. The system maps the overlap portions of the first hemispherical image and the second hemispherical image to the 2D projection of the cubic image, and encodes the 2D projection of the cubic image to generate an encoded image representative of the spherical FOV.

Mapping of spherical image data into rectangular faces for transport and decoding across networks

A spherical content capture system captures spherical video content. A spherical video sharing platform enables users to share the captured spherical content and enables users to access spherical content shared by other users. In one embodiment, captured metadata or video/audio processing is used to identify content relevant to a particular user based on time and location information. The platform can then generate an output video from one or more shared spherical content files relevant to the user. The output video may include a non-spherical reduced field of view such as those commonly associated with conventional camera systems. Particularly, relevant sub-frames having a reduced field of view may be extracted from each frame of spherical video to generate an output video that tracks a particular individual or object of interest.

Automatic generation of video from spherical content using audio/visual analysis

Generation of video from spherical content using edit maps

Automatic generation of video from spherical content using location-based metadata

A spherical content capture system captures spherical video and audio content. In one embodiment, captured metadata or video/audio processing is used to identify content relevant to a particular user based on time and location information. The platform can then generate an output video from one or more shared spherical content files relevant to the user. The output video may include a non-spherical reduced field of view such as those commonly associated with conventional camera systems. Particularly, relevant sub-frames having a reduced field of view may be extracted from each frame of spherical video to generate an output video that tracks a particular individual or object of interest. For each sub-frame, a corresponding portion of an audio track is generated that includes a directional audio signal having a directionality based on the selected sub-frame.

Automatic generation of video and directional audio from spherical content

A pair of cameras having an overlapping field of view is aligned based on images captured by image sensors of the pair of cameras. A pixel shift is identified between the images. Based on the identified pixel shift, a calibration is applied to one or both of the pair of cameras. To determine the pixel shift, the camera applies correlation methods including edge matching. Calibrating the pair of cameras may include adjusting a read window on an image sensor. The pixel shift can also be used to determine a time lag, which can be used to synchronize subsequent image captures.

Target-less auto-alignment of image sensors in a multi-camera system

Images captured by multi-camera arrays with overlap regions can be stitched together using image stitching operations. An image stitching operation can be selected for use in stitching images based on a number of factors. An image stitching operation can be selected based on a view window location of a user viewing the images to be stitched together. An image stitching operation can also be selected based on a type, priority, or depth of image features located within an overlap region. Finally, an image stitching operation can be selected based on a likelihood that a particular image stitching operation will produce visible artifacts. Once a stitching operation is selected, the images corresponding to the overlap region can be stitched using the stitching operation, and the stitched image can be stored for subsequent access.

Image stitching in a multi-camera array

Multiple cameras are arranged in an array at a pitch, roll, and yaw that allow the cameras to have adjacent fields of view such that each camera is pointed inward relative to the array. The read window of an image sensor of each camera in a multi-camera array can be adjusted to minimize the overlap between adjacent fields of view, to maximize the correlation within the overlapping portions of the fields of view, and to correct for manufacturing and assembly tolerances. Images from cameras in a multi-camera array with adjacent fields of view can be manipulated using low-power warping and cropping techniques, and can be taped together to form a final image.

Growing community of inventors

Woodside, CA, United States of America

Timothy MacMillan