Amazon Technologies, Inc.

Samsara Inc.

William Nathan John Hurst

Bradley Lawrence Hinkel

Timothy Daniel Cole

Christopher Gerald Emery

Ivan Avramov

Jonathan Peter Wright

William Evan Welbourne

Jing Wang

Greg Allen Cummings

Narendran Rajan

Christopher Mark Paola

Jason Chern Hooi Chionh

Siddharth Heroor

Methods, systems, and apparatuses capture event video segments from an IoT device such as a dashcam in a vehicle. When an even is detected, the original resolution video segment is converted to a low resolution video segment and stored in a cloud backend. Some embodiments thus save bandwidth and storage space by transmitting low resolution video segment to the cloud backend and further save bandwidth and storage space by only storing video in the cloud backend when an event is detected. Bandwidth is also saved because, in some embodiments, original resolution video segments are stored on the IoT device and only transmitted to a reviewer when requested. In the cloud backend, the low resolution video segment is converted to a higher resolution video using a deep learning video upscaler before it is reviewed by a reviewer. A human reviewer can also review the original resolution video segment that is stored on the IoT device if it is desired to see the video segment in its original resolution. In some embodiments, a reviewer can request that an area of interest in the video segment being reviewed is upscaled. In such embodiments the entire low resolution video segment is not always upscaled in its entirety before review.

Optimizing video data upload for dashcam event reviews

Documents may be maintained in a repository and retrieved based on searches that specify labels as criteria. Documents may be associated with groups of labels identified as topics. Searches may be performed using binary-encoded matrices specifying relationships between documents and topics, topics and labels and differential information indicating differences between topics and labels associated with documents. An initial result estimate may be based on forming a product of a documents-topics matrix and a topics-labels matrix. The initial estimate may be corrected by applying the differential information.

Approximation query

A device may participate in a shared data state in which editing operations performed locally are distributed and synchronized with other devices participating in the shared data state. Data may be maintained in log-structured storage files on each device. Changes made locally may be conditionally appended to log-structured storage and committed or rolled back during synchronization. Devices may exchange sets of related change descriptors preceded by associated contextual information. Conflicts may be resolved through application of conflict resolution rules or through a user-involved process. User notification of conflict resolution may be based on user intent inferred from contextual information and associated change descriptors.

Log-based synchronization with inferred context

A system and method for obtaining, by an electronic device, a first set of image data recorded during an image capture event and a second set of image data recorded during an image capture event. During the image capture event, a characteristic of an image capture device is changed from a first state to a second state. The first and second sets of image data are recorded by an image capture device having one or more image recording components with one or more image sensor arrays. An image is generated based at least in part on the first and second sets of image data.

Pixel readout of a charge coupled device having a variable aperture

Image filter values can be obtained by applying a high-frequency or edge-detection image filter to an image to extract a level of image detail. The image can be divided into blocks of a predetermined size and the image filter values of the pixels in a block can be used to obtain a density value associated with the level of image detail for the block. For blocks where the density value exceeds a threshold amount, a degree of blur may be applied based on the density value. Thus, the image can be rendered so that only some of the blocks of the image are blurred while other blocks do not have blur applied.

Processing an image to reduce rendering artifacts

Computing devices may synchronize respective copies of a repository in part by sharing status information for records stored in the repository. Status information may be conveyed by encoding record identifiers in a hashtable. A hash function may be selected to have an output range equal to or larger than a number of records in the repository. A plurality of hash function outputs may be mapped to a storage location containing a sum of identifiers and a count of identifiers included in the sum. The storage location may be transmitted to a remote computing device, which may unfold the data in the storage location to identify changed records.

Folded-hashtable synchronization mechanism

A system and method for determining information about a scene of an image capture event, wherein a first set of image data and a second set of image data of a scene is recorded during an image capture event. Information is determined about the scene by comparing the first and second sets of image data and an image is generated based on the determined information. In some embodiments, depth map information is further generated based on the information. In some embodiments, the image is deblurred based on the information.

Multiple captures with a variable aperture

An system having an aperture configured to change state while recording one or more images during an image capture process, and a method for processing one or more recorded images by detecting at least a bokeh artifact within the one or more images and processing the images based at least in part on information obtained by analyzing the size and orientation of the bokeh artifact.

Variable temporal aperture

A device may participate in a shared data state in which editing operations performed locally are distributed and synchronized with other devices participating in the shared data state. Data may be maintained in log-structured storage files on each device. Changes made locally may be represented by change descriptors conditionally appended to the end of a log. The size of the log may be monotonically increasing. Monotonically increasing identifiers may be used to implicitly or explicitly identify change descriptors. Upon conflict resolution, conditionally appended change descriptors may be committed. Non-committed change descriptors may be conditionally reappended to the end of the log for further conflict resolution.

Log-based synchronization with conditional append

The following description is directed to systems and methods for digital cameras. In one embodiment, a camera can comprise an optical system, a sensor array, a color spreader, a memory, and a controller. The optical system can focus an image. The sensor array comprises a plurality of pixels and can be configured to capture image data. The color spreader can be positioned between the optical system and the sensor array. The color spreader can comprise an array of segments, where each respective segment can be configured to spread a spectrum of incident light toward a corresponding group of pixels using diffraction or refraction. The memory can store one or more spatial models. The controller can be coupled to the sensor array and the memory. The controller can be configured to reconstruct spectral and spatial components of the image from the captured image data and one of the spatial models.

Digital camera for capturing spectral and spatial information

A device may participate in a shared data state in which editing operations performed locally are distributed and synchronized with other devices participating in the shared data state. Data may be maintained in log-structured storage files on each device. Changes made locally may be conditionally appended to log-structured storage and committed or rolled back during synchronization. Sets of related change descriptors may be identified based on inferred user intent. Conflict resolution rules may be applied to sets of related change descriptors. Conflicts may be resolved through application of conflict resolution rules or through a user-involved process.

Growing community of inventors

Seattle, WA, United States of America

William Nathan John Hurst