Ambarella International LP

Ruian Xu

Jian Tang

Zhe Zhang

Zhikan Yang

Christopher N Day

Luyi Sun

Didier LeGall

Long Zhao

Udit Budhia

Peichun Peng

An apparatus comprising an infrared (IR) illuminator, an interface and a processor. The IR illuminator may comprise an array of emitters divided into a plurality of segments each configured to generate an amount of an IR light. The interface may be configured to receive pixel data comprising the IR light. The processor may be configured to process the pixel data arranged as video frames, extract IR illumination data for each of a plurality of zones of the video frames, perform a comparison of the IR illumination data in each zone to an exposure threshold, and generate a control signal in response to the comparison. Each zone may correspond to a subsection of the video frames associated with a respective one of the segments. The control signal may be configured to independently adjust the amount of the IR light generated by each of the segments for each of the zones.

IR illumination control for cameras with multi-region array

An apparatus comprising an interface, a sensor and a processor. The interface may receive pixel data of an exterior of a vehicle. The sensor may detect an obstacle at a distance from the vehicle. The processor may process the pixel data arranged as video frames, perform computer vision operations on the video frames to detect objects, determine whether one of the objects is a vehicle attachment, determine an amount of distance that the vehicle attachment extends a size of the vehicle and generate a recalibration value for the distance used by the sensor in response to the amount of distance. A notification may be generated in response to the sensor detecting the obstacle within the distance. The distance may be a default value based on the size of the vehicle when the vehicle attachment is not present and the recalibration value when the vehicle attachment is detected.

Dynamic adjustment of sensor range using computer vision

An apparatus comprising an interface and a processor. The interface may be configured to receive pixel data corresponding to a monitored area. The processor may be coupled to the interface and may be configured to (i) process the pixel data arranged as video frames, (ii) perform computer vision operations to detect an object in the video frames, (iii) extract characteristics about the object detected, (iv) generate a three-dimensional (3D) map of the object utilizing the pixel data corresponding to the object, (v) determine whether the object is within a three-dimensional (3D) virtual security zone based on the 3D map of the object, and (vi) generate a trigger signal in response to all or a portion of the object being within the 3D virtual security zone. The 3D virtual security zone is generally determined by a user during a setup procedure.

Security cameras integrating 3D sensing for virtual security zone

An apparatus includes a first light source, a second light source, an image sensor circuit, and a processing circuit. The first light source is generally capable of emitting infrared light. The second light source is generally capable of emitting visible light. The image sensor circuit is generally responsive to both the infrared light and the visible light. The processing circuit is generally coupled to the image sensor circuit and configured to generate an image comprising both infrared information and color information.

IP security camera combining both infrared and visible light illumination plus sensor fusion to achieve color imaging in zero and low light situations

An apparatus including an interface and a processor. The interface may be configured to receive pixel data generated by a capture device. The processor may be configured perform computer vision operations on the video frames to detect objects, perform a classification of the objects detected based on characteristics of the objects, determine whether the classification of the objects corresponds to an event, generate a full video stream in response to all of the video frames and generate encoded video frames. The full video stream may be recorded to a storage medium local to the apparatus. The encoded video frames may be communicated to a cloud service. The encoded video frames may comprise a first sample of the video frames selected at a first rate when the event is not detected and a second sample of the video frames selected at a second rate while the event is detected.

Smart timelapse video to conserve bandwidth by reducing bit rate of video on a camera device with the assistance of neural network input

An apparatus comprising an interface and a processor. The interface may be configured to receive pixel data generated by a capture device and an activation signal generated by at least one of a plurality of sensors. The processor may be configured to process the pixel data arranged as video frames, perform computer vision on the video frames to detect objects, perform a classification of the objects, present a control signal in response to receiving the activation signal, determine whether there is a match between the classification and an object class and generate a calibration signal in response to the match. The computer vision may determine which zone that the objects have been detected in. Each zone may comprise a coverage region of one of the sensors. The calibration signal may adjust parameters of the sensors that correspond to the zone with the objects that do not have the match.

Dynamically adjusting activation sensor parameters on security cameras using computer vision

An apparatus includes a camera and a processing circuit. The camera may be configured to capture images of an environment around a vehicle. The processing circuit may be configured to (i) perform automated number-plate recognition using the images, (ii) store a history of detected license plates, and (iii) search the history of detected license plates in response to receiving a request from a communication device of a vehicle user for information matching the request.

Advanced number plate recognition implemented on dashcams for automated amber alert vehicle detection

An apparatus includes an interface circuit and a control circuit. The interface circuit may be configured to receive pixel data corresponding to a field of view of a camera. The control circuit may be configured to process the pixel data arranged as video frames and control an exposure time for capturing the pixel data and a turn on time of a structured light pattern to obtain a sequence of images comprising at least one image including the structured light pattern and at least one image where the structured light pattern is absent. The control circuit may be further configured to perform a depth analysis to generate depth information using the at least one image including the structured light pattern. The control circuit may store and execute an artificial neural network trained to (i) discern whether a face is at least one of a real face and a fake face, and (ii) make a liveness determination utilizing the depth information.

Timing mechanism to derive non-contaminated video stream using RGB-IR sensor with structured light

An apparatus includes an interface and a processor. The interface may be configured to receive pixel data corresponding to an entrance of a secured area. The processor may be configured to process the pixel data arranged as video frames, perform computer vision operations to detect objects, extract characteristics about the objects, determine a number of people in the video frames, receive an authorization signal from an access control device and generate a control signal. The authorization signal may be used by the access control device to grant access to the secured area. The number of people in the video frames may be determined after the authorization signal is received. The control signal may be generated when the number of people is greater than a number of entrants permitted by the authorization signal. The control signal may provide an indication of unauthorized access to the secured area.

Sensor for access control reader for anti-tailgating applications

An apparatus including an interface and a processor. The interface may be configured to receive pixel data. The processor may be configured to generate a plurality of video frames in response to the pixel data received from the interface, perform computer vision operations to detect objects in the video frames, extract features of the objects in response to characteristics of the objects determined using the computer vision operations, identify a person in the video frames based on the features, detect an event based on the person identified and generate a notification in response to detecting the event and a permission status. The permission status may suppress sending the notification when the permission status for the person identified corresponds to denying the notification and enable sending the notification when the permission status does not correspond to denying the notification.

Computer vision enabled smart snooze home security cameras

An apparatus including a capture device and one or more sensors. The capture device may be configured to operate in a standby mode of operation until a wakeup signal is received and operate in video capture mode of operation in response to the wakeup signal. The sensors may each be configured to detect a trigger condition and communicate the wakeup signal in response to detecting the trigger condition. The capture device may be configured to consume a first amount of power in the standby mode of operation and a second amount of power in the video capture mode of operation. The first amount of power may be less than the second amount of power. The capture device may be configured to generate pixel data in the video capture mode of operation. Each of the sensors may be remotely located on a vehicle with respect to the capture device.

Car security camera triggering mechanism

An apparatus including an interface and a processor. The interface may be configured to receive pixel data generated by a capture device. The processor may be configured to generate video frames in response to the pixel data, perform computer vision operations on the video frames to detect objects, perform a classification of the objects detected based on characteristics of the objects, determine whether the classification of the objects corresponds to a user-defined event and generate encoded video frames from the video frames. The encoded video frames may be communicated to a cloud storage service. The encoded video frames may comprise a first sample of the video frames selected at a first rate when the user-defined event is not detected and a second sample of the video frames selected at a second rate while the user-defined event is detected. The second rate may be greater than the first rate.

Event/object-of-interest centric timelapse video generation on camera device with the assistance of neural network input

An apparatus comprises a camera and one or more sensors. The camera generally has a low power deactivated mode. The one or more sensors are generally remotely located with respect to the camera. The one or more sensors may be configured to communicate a signal to the camera in response to a trigger condition. The camera is generally configured to activate in response to receiving the signal from the one or more sensors.

Growing community of inventors

San Jose, CA, United States of America

Ruian Xu