Abstract: Provided is a ranging sensor (2), including: a light emitting unit (110) that applies light to a target; a light receiving unit (120) that receives light from the light emitting unit (110) reflected from the target; and a calculation unit (150) that calculates a distance from the light receiving unit (120) to the target based on light reception data acquired by the light receiving unit (120). The calculation unit (150) compares detection light reception data with predetermined data, the detection light reception data being obtained as light is emitted from the light emitting unit (110) to a reference object provided at a predetermined distance and the light of the light emitting unit (110) reflected by the reference object is received by the light receiving unit (120), and controls notification processing to a user.

Abstract: In one embodiment, a computing system may transmit, using one or more light emitters, light beams of different wavelengths simultaneously into a surrounding environment. The system may determine a characteristic of the surrounding environment based on reflections of the light beams. In response to a determinization that the characteristic of the surrounding environment satisfies a criterion, the system may configure the one or more light emitters to transmit light beams of different wavelengths sequentially into the surrounding environment for measuring distances to one or more objects in the surrounding environment.

Abstract: The sensory substitution system includes a portable matrix of exciters that can be worn by a user in such a way that the exciters are placed against the skin of a user, as well as a control circuit for the exciters of the matrix. An asynchronous signal source provides the control circuit with an asynchronous signal representative of visual information organised according to a matrix of pixels. This asynchronous signal includes, for each pixel, successive events associated in an asynchronous manner with the pixel.

Abstract: An approach is provided for providing predictive classification of sensor error. The approach involves, for example, receiving sensor data from at least one sensor, the sensor data collected at a geographic location. The approach also involves extracting a set of input features from the sensor data, map data representing the geographic location, or combination thereof. The approach further involves processing the set of input features using a machine learning model to calculate a predicted sensor error of a target sensor operating at the geographic location. The machine learning model, for instance, has been trained on ground truth sensor error data to use the set of input features to calculate the predicted sensor error.

Abstract: Sensors coupled to a vehicle are calibrated using a dynamic scene with sensor targets around a motorized turntable that rotates the vehicle to different orientations. The sensors capture data at each orientation along the rotation. The vehicle's computer identifies representations of the sensor targets within the data captured by the sensors, and calibrates the sensor based on these representations. The motorized turntable may confirm that rotation has stopped to the vehicle to trigger sensor capture, and the vehicle may communicate completion of sensor capture at an orientation to the motorized turntable to trigger further rotation.

Abstract: Systems and techniques are described for processing media content, such as samples of sparse media content whose sample durations can be unknown at the time of decoding the samples, include obtaining, at a current time instance, a current fragment comprising at least a current time component. From the current time component, a modified duration is obtained for at least one media sample, the modified duration indicating a time duration by which a presentation of a previous media sample of a previous fragment is to be extended or reduced in relation to the current time instance. The at least one media sample is presented for a time duration based on the modified duration.

Abstract: A system comprising a coordinate tracking engine and a video classification engine communicably coupled to a notification engine. The coordinate tracking engine detects that geographical coordinates of a mobile device indicate that an account holder is within a threshold distance of a physical branch of an institution. The notification engine retrieves account information for the account holder. The coordinate tracking engine further detects that the account holder has arrived at the physical branch. The video classification engine captures video frames of an entrance to the physical branch and identifies the account holder. The notification engine further presents account information for the account holder on a display.

Abstract: The present disclosure advantageously provides a heterogenous system, and a method for generating an artificial neural network (ANN) for a heterogenous system. The heterogenous system includes a plurality of processing units coupled to a memory configured to store an input volume. The plurality of processing units includes first and second processing units. The first processing unit includes a first processor and is configured to execute a first ANN, and the second processing unit includes a second processor and is configured to execute a second ANN. The first and second ANNs respectively include an input layer, at least one processor-optimized hidden layer and an output layer. The second ANN hidden layers are different than the first ANN hidden layers.

Abstract: Methods and devices are provided for adding lyrics to a short video. The device obtains a music material required by the short video and obtains a first playback duration of the short video. The device obtains a target music material having a playback duration matching the first playback duration. The device obtains a lyric sticker corresponding to the target music material based on the lyrics extracted from the target music material and displays a processed short video after adding with the lyric sticker.

Abstract: Indexing of media sources available to media composition applications, such as video editing applications and digital audio workstations, is extended to sources of an arbitrary type, including non-file type sources. Examples of such sources include devices using baseband video and audio protocols, network ports and physical devices connected via USB, Thunderbolt, etc. The sources are discovered and indexed even when not previously known to the application with the result that any available bitstream becomes available to the application. Applications access the sources via feeders, which are addressable portals through which media essence is received. The indexing methods involve the instantiation of a hierarchy of plug-in software modules by a format-handling subsystem of the application or by a media indexing service servicing one or more applications.

Abstract: A vehicular test system for testing a vehicular sensing system includes a sensor support structure having a proximal end disposed at a vehicle, a distal end extending away from the vehicle, and a force providing element that provides a force to move the distal end of the sensor support structure. A vehicular sensor is disposed at the distal end of the sensor support structure. When the vehicular sensor is approaching a collision with an object, such as during testing of vehicular sensors and vehicular sensing systems, a control controls the force providing element to move the distal end of the sensor support structure and the vehicular sensor to avoid the collision.

Abstract: An imaging system including a front aperture, two or more refractive lens elements mounted in a lens barrel, and a photosensor. One or more of the components of the imaging system (e.g., the aperture, lenses, lens groups, and/or photosensor) are tilted with respect to each other and/or with respect to a center (or mechanical) axis of the imaging system to compensate for effects including but not limited to keystone distortion, resolution non-uniformity, and gradient blur that result from tilt of an object in the field of view of the camera with respect to the center axis of the camera.

Abstract: A system and method for testing a Light Detection and Ranging (LiDAR) sensor is disclosed. The system includes the LiDAR sensor radiating a plurality of LiDAR rays, a plurality of bars positioned at a predetermined distance from the LiDAR sensor, and a testing device. The method includes triggering a LiDAR sensor to radiate a plurality of LiDAR rays. The method further includes determining at least one intersection point at each of a plurality of bars upon intersection of at least one LiDAR ray from the plurality of LiDAR rays with each of the plurality of bars. The method further includes computing at least one operational parameter for the LiDAR sensor based on the intersection points, and determining one or more test results based on the one or more operational parameters.

Abstract: Systems and methods are provided for reordering and/or bypassing certain informational content or menus that are conventionally presented prior to playback of media content stored on physical media discs. Upon initial use of a physical media disc, certain information content or menus may be presented to a user or viewer, for example, piracy warnings, language selection menus, etc. However, upon subsequent use of the physical media disc, such informational content or menus may be bypassed. The user or viewer is given an option to immediately begin consuming the media content stored on the physical media disc. Conventional content, such as trailers are not played prior to playback of the media content.

Abstract: A system and method for alerting a driver of a motor vehicle or a person walking along a road or hiking on a trail of potentially dangerous hazards in their path. Hazards may be deep water, ice, oil slicks or other hazards. In the case of a motor vehicle, the system uses cameras mounted on or within the vehicle to detect potential hazards and then analyzes the images combined with the known topography of the location to evaluate the ability of the vehicle to safely traverse the hazard. In the case of a person walking or hiking, the person may use the camera on a personal mobile device to capture images of the hazard and to combine the images with the known topography at the location to evaluate the danger presented by the hazard.

Abstract: Example embodiments provide systems and methods for replacing recorded media content. In an example method, at least one media content source is searched for a media content item recorded to storage of a first media content device. Based on the media content item being found at a first media content source of the at least one media content source, a recording of the media content item to storage of a second media content device is scheduled based on the media content item being recordable from the first media content source at the second media content device. Based on the media content item not being recordable from the first media content source at the second media device, the availability of the media content item is indicated to the second media content device.

Abstract: Aspects of the present disclosure enable humanly-specified relationships to contribute to a mapping that enables compression of the output structure of a machine-learned model. An exponential model such as a maximum entropy model can leverage a machine-learned embedding and the mapping to produce a classification output. In such fashion, the feature discovery capabilities of machine-learned models (e.g., deep networks) can be synergistically combined with relationships developed based on human understanding of the structural nature of the problem to be solved, thereby enabling compression of model output structures without significant loss of accuracy. These compressed models provide improved applicability to “on device” or other resource-constrained scenarios.

Abstract: Techniques for training a machine learning model are described herein. For example, the techniques may include implementing a cross batch normalization layer that generates a cross batch normalization layer output based on a first layer output during training of the neural network. The training may be based on a local batch of training examples of a global batch including the local batch and at least one remote batch of training examples. The cross batch normalization layer output may include normalized components of the first layer output determined based on global normalization statistics for the global batch. Such techniques may be used to train a neural network over distributed machines by synchronizing batches between such machines.

Abstract: A multimedia file and methods of generating, distributing and using the multimedia file are described. Multimedia files in accordance with embodiments of the present invention can contain multiple video tracks, multiple audio tracks, multiple subtitle tracks, data that can be used to generate a menu interface to access the contents of the file and ‘meta data’ concerning the contents of the file. Multimedia files in accordance with several embodiments of the present invention also include references to video tracks, audio tracks, subtitle tracks and ‘meta data’ external to the file. One embodiment of a multimedia file in accordance with the present invention includes a series of encoded video frames and encoded menu information.

Abstract: Embodiments of the present application provide a video recording method and device. The method includes: receiving, by an application of a mobile terminal, a video recording instruction of a user; generating, according to a photographing scene, a video stream that can be edited according to segments, the video stream including at least one frame sequence segment; and uploading the edited video stream to a server. According to the present application, by generating a video stream consisting of at least one frame sequence segment during video recording, where the generated video stream can be edited according to segments, the error tolerance of video capture is improved, and the costs of later video editing are reduced.