Samsara Inc.

Fuji Electric Co., Ltd.

Tapplock Corporation

Jing Wang

Sharan Srinivasan

John Charles Bicket

Brian Tuan

Muhammad Ali Akhtar

Bruce Kellerman

Vincent Shieh

Abner Ayala Acevedo

Margaret Irene Finch

Yan Wang

William Evan Welbourne

Nathan Hurst

Akshay Raj Dhamija

Rohit Annigeri

Howard Yu

Shaurye Agarwal

Tetsuya Kawashima

Masahiro Sasaki

Wai Tung Ng

Ingo Gerhard Wiegand

Greg Allen Cummings

William Nathan John Hurst

Haruhiko Nishio

Shirish Nair

Narendran Rajan

Kendy Ng

Ran Tao

Jing Hua Ye

Weijie Li

Siddharth Heroor

Faraz Abbasi

The present disclosure is related to systems and methods of dashcam installation for providing instructions on placing a dashcam for a person or dashcam installer. Aspects of the present disclosure are related to providing a digital assistant for installing the dashcam in an acceptable position, where the dashcam can capture images of objects with acceptable quality (e.g., as determined by a machine learning model). In some embodiments, the dashcam detects a driver and provides inferences related to the current dashcam position. The inferences can be based on analyzing images captured by the dashcam in the current position, where the analysis can utilize a machine learning model. The inferences can also provide one or more instructions to move and/or tilt the dashcam to the acceptable installation position.

Systems and methods for dashcam installation

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

Methods, systems, and programs are presented for detecting impaired views in monitoring cameras. One method includes training a rotation classifier with unsupervised learning utilizing a first set of images. The rotation classifier is configured to receive an input image and generate a rotation feature embedding for the input image. In addition, the method includes training an impairment classifier with supervised learning utilizing a second set of images, impairment labels for each of the second set of images, and the rotation feature embedding, generated by the rotation classifier, for each of the second set of images. The method further includes accessing a vehicle image captured by a camera on a vehicle, and providing the vehicle image to the impairment classifier as input, and the impairment classifier outputs a camera impairment from a set of camera impairment categories. Further, the vehicle image and the camera impairment are presented on a user interface.

Detection of camera with impaired view

A vehicle device may execute one or more neural networks (and/or other artificial intelligence), such as based on input from one or more of the cameras and/or other sensors associated with the dash cam, to intelligently detect safety events in real-time. The vehicle device may further pass the input to a backend server for further analysis and the backend server can detect safety events based on the input. The vehicle device may analyze the output of the vehicle device and the output of the backend server to determine whether the output of the vehicle device is correct. If the output of the vehicle device is incorrect, the vehicle device can adjust how the vehicle device identifies safety events.

Refining event triggers using machine learning model feedback

A vehicle device may execute one or more neural networks (and/or other artificial intelligence), based on input from one or more of the cameras and/or other sensors, to intelligently detect safety events in real-time. The one or more neural networks may be an ensemble neural network that includes neural networks for detecting a head and hand of a user, neural networks for detecting hand actions of the user, neural networks for detecting the head pose of the user, neural networks for predicting an occurrence of an event, and neural networks for predicting a start time and end time of the event. Further, the neural networks can be segmented into a modular neural network based on metadata. The segmentation of the neural network can define a thin layer of the modular neural network to enable independent tuning of the thin layer of the modular neural network.

Ensemble neural network state machine for detecting distractions

Tuning layers of a modular neural network

A padlock device includes a housing; a shackle associated within the housing and having, with respect to the housing, a closed configuration and an open configuration; a latch subsystem associated with the housing for securely retaining the shackle in the closed configuration, the latch subsystem electrically operable to release the shackle; a biometric sensor associated with the housing to electronically sense fingerprint data from a finger being sensed; a control subsystem in the housing in communication with the biometric sensor and the latch subsystem, the control subsystem comprising: internal processor-readable memory configured to store one or more fingerprint records, each fingerprint record comprising authorized fingerprint data associated with a respective fingerprint identifier; processing structure configured to receive sensed fingerprint data from the biometric sensor and to cause the latch subsystem to release the shackle in the event of a release condition requiring at least that the sensed fingerprint data corresponds to authorized fingerprint data in at least one of the fingerprint records; the processing structure configured to present a management interface accessible by an external device in authorized communication with the control system to selectively: store one or more fingerprint records in the internal processor-readable memory; and delete or disable one or more stored fingerprint records in the internal processor-readable memory based at least on one or more respective fingerprint identifiers provided by the external device. A padlock system includes the padlock device; and a processor-readable medium embodying a computer program for provisioning an external device to conduct authorized communications with the padlock device, the computer program including program code for presenting a user interface on the external device for enabling the authorized manager to conduct managing of fingerprint records for the padlock device; and program code for accessing the management interface of the padlock device in accordance with the managing.

Padlock device, systems including a padlock device, and methods of operating therefor

A fully integrated DC-DC converter utilizes digitally controlled dual output stages to achieve fast load transient recovery is presented. The DC-DC converter includes a main converter output stage connected in parallel with an auxiliary output stage. The main output stage is responsible for steady-state operation and is designed to achieve high conversion efficiency using large inductor and power transistors with low on-resistance. The auxiliary stage is responsible for transient suppression and is only active when a load transient occurs. The auxiliary output stage performs well with inductor and power transistors much smaller than those of the main switching stage and thus achieves well balanced power conversion efficiency and dynamic performance with a much smaller area penalty than previously described dual-output-stage converters.

Digitally controlled integrated DC-DC converter with transient suppression

Growing community of inventors

Toronto, Canada

Jing Wang