Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing instance segmentation by detecting and segmenting individual objects in an image. In one aspect, a method comprises: processing an image to generate data identifying a region of the image that depicts a particular object; obtaining data defining a plurality of example object segmentations; generating a respective weight value for each of the example object segmentations; for each of a plurality of pixels in the region of the image, determining a score characterizing a likelihood that the pixel is included in the particular object depicted in the region of the image using: (i) the example object segmentations, and (ii) the weight values for the example object segmentations; and generating a segmentation of the particular object depicted in the region of the image using the scores for the pixels in the region of the image.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: Provided are systems and methods that invert a trained NeRF model, which stores the structure of a scene or object, to estimate the 6D pose from an image taken with a novel view. 6D pose estimation has a wide range of applications, including visual localization and object pose estimation for robot manipulation.

Abstract: A computer system for failing a secure boot in a case tampering event comprises a microcontroller unit (MCU); a trusted platform module (TPM), for generating random bytes for a secure boot of the computer system; a bootloader, for storing information comprising the random bytes in the MCU and at least one hardware of the computer system and performing the secure boot, wherein the TPM is comprised in the bootloader; an operating system (OS), for performing the secure boot; and at least one sensor, coupled to the MCU, for detecting a case tampering event, and transmitting a signal for triggering a deletion of the random bytes, if the case tampering event happens. The MCU performs the operation of deleting the random bytes stored in the MCU and the at least one hardware according to a power supply, in response to the signal.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: An electronic device is provided. The electronic device includes an ultrasonic input module and a control unit. The control unit is electrically connected to the ultrasonic input module. The control unit is configured to: continuously read a setting signal of the ultrasonic input module within a preset time to generate a threshold, where the setting signal includes a plurality of sensing values; receive an input signal by using the ultrasonic input module, where the input signal includes a feature value; and compare the feature value with the threshold to generate associated information, and process the input signal according to the associated information. The disclosure further provides a touch method applied to the electronic device.

Abstract: The present technology relates generally to the preparation of immunoglobulin-related compositions (e.g., antibodies or antigen binding fragments thereof) that specifically bind STEAP1 protein and uses of the same. In particular, the present technology relates to the preparation of STEAP1 binding antibodies and their use in detecting and treating STEAP1-associated cancers.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training an image representation neural network.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: Changing a lighting mode for a human interface device is described herein. A first lighting mode can be initiated for a human interface device. Keys on the human interface device can be selected over a period of time at a frequency that is within a defined range. A second lighting mode for the human interface device can be identified based in part on the frequency being within the defined range. The first lighting mode and the second lighting mode can define a lighting scheme for light sources in the human interface device that reflect a user mood. The first lighting mode can be switched to the second lighting mode.

Abstract: An electronic device is provided. The electronic device includes an ultrasonic input module and a control unit. The control unit is electrically connected to the ultrasonic input module. The control unit is configured to: continuously read a setting signal of the ultrasonic input module within a preset time to generate a threshold, where the setting signal includes a plurality of sensing values; receive an input signal by using the ultrasonic input module, where the input signal includes a feature value; and compare the feature value with the threshold to generate associated information, and process the input signal according to the associated information. The disclosure further provides a touch method applied to the electronic device.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: A computer-implemented method of generating scale-permuted models can generate models having improved accuracy and reduced evaluation computational requirements. The method can include defining, by a computing system including one or more computing devices, a search space including a plurality of candidate permutations of a plurality of candidate feature blocks, each of the plurality of candidate feature blocks having a respective scale. The method can include performing, by the computing system, a plurality of search iterations by a search algorithm to select a scale-permuted model from the search space, the scale-permuted model based at least in part on a candidate permutation of the plurality of candidate permutations.

Abstract: A computing system for producing an architecture of a pyramid layer is disclosed. The computing system can include a controller model configured to generate new architectures for a pyramid layer that receives a plurality of input feature representations output by a backbone model and, in response, outputs a plurality of output feature representations. The plurality of input feature representations can have a plurality of different input resolutions, and the plurality of output feature representations can have a plurality of different output resolutions. The computing system can be configured to perform a plurality of iterations. For each iteration, the computing system can receive a new pyramid layer architecture as an output of the controller model and evaluate one or more performance characteristics of a machine-learned pyramidal feature model that includes the backbone model and one or more pyramid layers that have the new pyramid layer architecture.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing instance segmentation by detecting and segmenting individual objects in an image. In one aspect, a method comprises: processing an image to generate data identifying a region of the image that depicts a particular object; obtaining data defining a plurality of example object segmentations; generating a respective weight value for each of the example object segmentations; for each of a plurality of pixels in the region of the image, determining a score characterizing a likelihood that the pixel is included in the particular object depicted in the region of the image using: (i) the example object segmentations, and (ii) the weight values for the example object segmentations; and generating a segmentation of the particular object depicted in the region of the image using the scores for the pixels in the region of the image.

Abstract: A communication system includes a management device of Internet of Thing (IoT) devices and an agent communicatively coupled between the management device of IoT devices and plural cloud servers. The agent receives identification information corresponding to a first IoT device, and obtains a key hash corresponding to the first IoT device from one of the cloud servers that corresponds to the first IoT device according to the identification information, and transmits the key hash to the management device of IoT devices. The management device of IoT devices encrypts management information by utilizing the key hash and transmits the encrypted management information to the first IoT device. The management information includes a control hash. The first IoT device decrypts the management information to obtain the control hash, and encrypts a message by utilizing the control hash, and transmits the encrypted message to the management device of IoT devices.

Abstract: A micro-light-emitting diode device includes a first semiconductor layer, an active layer, and a second semiconductor layer. The first semiconductor layer has a first bottom surface. The active layer is disposed on the first semiconductor layer. The second semiconductor layer is disposed on the active layer. The second semiconductor layer and the active layer have an interface. A surface of the second semiconductor layer opposite to the active layer is a light-exiting surface of the micro-light-emitting diode device. A distance between the light-exiting surface and the interface decreases from a central axis of the second semiconductor layer to an edge of the second semiconductor layer, so as to provide a focusing effect for the light by the light-exiting surface.

Abstract: A detecting device includes a turntable, a rotating device, a sound generating device, and a processing device. The turntable has a first surface and a second surface opposite to the first surface, and the first surface is used to dispose a device to be tested. The rotating device is connected to the second surface of the turntable for driving to rotate the turntable. The sound generating device is used to emit a test audio toward the device to be tested. The processing device is connected to the rotating device and the sound generating device. Among them, the processing device controls a rotational speed of the rotating device to rotate the device to be tested, and controls the sound generating device to emit the test audio to the device to be tested such that the device to be tested may receive the test audio at different angles.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: A detecting device includes a turntable, a rotating device, a sound generating device, and a processing device. The turntable has a first surface and a second surface opposite to the first surface, and the first surface is used to dispose a device to be tested. The rotating device is connected to the second surface of the turntable for driving to rotate the turntable. The sound generating device is used to emit a test audio toward the device to be tested. The processing device is connected to the rotating device and the sound generating device. Among them, the processing device controls a rotational speed of the rotating device to rotate the device to be tested, and controls the sound generating device to emit the test audio to the device to be tested such that the device to be tested may receive the test audio at different angles.