Brainoft Inc.

Leland Stanford Junior University

Ashutosh Saxena

Deng Deng

Brendan Berman

Lukas Kroc

Maurice Kyojin Chu

Chenxia Wu

Jinjing Zhou

Ozan Sener

Andrew Y Ng

Hema Swetha Koppula

Ashwini Venkatesh

Yu Zhou

Siddharth Batra

Min Sun

Kunal Lad

Sung Hwan Chung

Sensor information is received. Based at least in part on at least one state detected using the sensor information, an expected value for automatically performing an action associated with a device controlled by a device controller system is evaluated as compared to an expected value for potentially performing, subject to a user interactive qualification, the action associated with the device controlled by the device controller. It is determined that the expected value of automatically performing the action is insufficient to trigger automatically performing the action without being subject to the user interactive qualification. A confirmatory indication is identified. Based at least in part on the confirmatory indication, an instruction is sent to the device to perform the action.

Interactive environmental controller

A first input is received from at least one of a plurality of sensors. Based on the first input, a first probability associated with a first state is determined. A second input is received from at least one of the plurality of sensors. Based on the second input, a second probability associated with a second state is determined. The second state is based on a transition model and the second probability. The transition model associates the first state with the second state and indicates a likelihood of a transition from the first state to the second state. A rule is triggered to change a state of at least one network connected device based on the second state.

Automatically learning and controlling connected devices

One embodiment can provide a light switch module for controlling at least one light source coupled to the light switch module. The light switch module can include a manual control configured to receive a user manual operation; a control-signal-receiving module configured to receive, from a central controller, a control signal; and a switch-mode-configuration module coupled to the manual control and the control-signal-receiving module. The switch-mode-configuration module is configured to place the light switch in a first operation mode in response to receiving the control signal and place the light switch in a second operation mode in response to not receiving the control signal within a predetermined time interval.

Custom universal light switch and dimmer

Sensor data is received from a physical environment sensor. A component region associated with at least a portion of the sensor data is identified. A physical environment has been defined to include a plurality of component regions. Context information associated with the identified component region is obtained. The context information is utilized in association with a prediction model.

Utilizing context information of environment component regions for event/activity prediction

One embodiment can provide a home-automation system. The home-automation device can include a mechanical supporting structure and a plurality of components mechanically coupled to the mechanical supporting structure. The plurality of components can include a central controller, one or more printed circuit boards (PCBs), and a plurality of sensors mounted on the one or more PCBs. The central controller can be configured to detect a user event based on output from multiple sensors, with each sensor being configured to sense parameters in a different physical domain.

Multi-modal interactive home-automation system

A network connected controllable device is automatically controlled. A plurality of inputs that indicate states of a network connected device environment associated with a property of the network connected controllable device to be automatically controlled is received. A plurality of output candidates is evaluated according to an optimization function. The optimization function depends on the plurality of inputs and one or more parameters and the output candidates are associated with candidate settings of the property of the network connected controllable device to be automatically controlled. An output that optimizes the optimization function is selected among the plurality of output candidates.

Controlling connected devices using an optimization function

Network connected devices are controlled. A command is received. A graph model is applied to identify a subset of nodes related to the command. The graph model includes a plurality of nodes that each correspond to a device and the graph model includes a plurality of edges that specify relationships between the plurality of nodes. The subset of nodes is activated in response to the command.

Controlling connected devices using a relationship graph

A first input is received from a plurality of sensors. A first state including a first location based on the first input is determined. The first state is associated with a first probability. A second input is received from the plurality of sensors. A second state including a second location is determined based on the second input associated with a second probability. It is determined that the second state corresponds to an actual state based on a transition model and the second probability. The transition model associates the first state with the second state and indicates a likelihood of a transition from the first state to the second state. A rule to change a state of at least one network connected device is triggered based on the second state.

Methods, systems, products and devices are implemented for editing video image frames. According to one such method, image content is embedded into video. A selection input is received for a candidate location in a video frame of the video. The candidate location is traced in subsequent video frames of the video by approximating three-dimensional camera motion between two frames using a model that compensates for camera rotations, camera translations and zooming, and by optimizing the approximation using statistical modeling of three-dimensional camera motion between video frames. Image content is embedded in the candidate location in the subsequent video frames of the video based upon the tracking thereof.

Systems, methods and devices for augmenting video content

Three-dimensional image data is generated. According to an example embodiment, three-dimensional depth information is estimated from a still image. A set of monocular images and their corresponding ground-truth depth maps are used to determine a relationship between monocular image features and the depth of image points. For different points in a particular image, the determined relationship is used together with local and global image features including monocular cues to determine relative depths of the points.

Growing community of inventors

Redwood City, CA, United States of America

Ashutosh Saxena