Abstract: This disclosure provides methods, devices, and systems for low-light imaging. The present implementations more specifically relate to selecting images that can be used for training a neural network to infer denoised representations of images captured in low light conditions. In some aspects, a machine learning system may obtain a series of images of a given scene, where each of the images is associated with a different SNR (representing a unique combination of exposure and gain settings). The machine learning system may identify a number of saturated pixels in each image and classify each of the images as a saturated image or a non-saturated image based on the number of saturated pixels. The machine learning system may then select the non-saturated image with the highest SNR as the ground truth image, and the non-saturated images with lower SNRs as the input images, to be used for training the neural network.

Abstract: This disclosure provides methods, devices, and systems for controlling motion-activated switches. The present implementations more specifically relate to relay controller that prevent motion-activated switches from turning off devices associated with an environment in which people are present. In some aspects, a motion-activated switch may include a relay controller coupled to a motion sensor, a camera, and a relay. The motion sensor outputs a motion trigger to the relay controller responsive to detecting motion in an environment. The motion trigger may cause the relay controller to acquire one or more images of the environment, via the camera, and selectively toggle the relay based on the acquired images. For example, the relay controller may close the relay responsive to identifying an image that includes an object of interest or may open the relay controller may open the relay responsive to identifying an image that does not include an object of interest.

Abstract: A display device includes a display panel, a backlight module, and backlight control circuitry. The backlight module is configured to illuminate the display panel, the backlight module comprising a plurality of light sources. The backlight control circuitry is configured to control first luminance of a first light source of the plurality of light sources based at least in part on a first weighted average of brightness values for a first set of pixels of the display panel.

Abstract: A voltage clamp is disclosed. The voltage clamp may include a plurality of transistors to limit the voltage between a power supply and ground. In addition, the voltage clamp may include a positive feedback signal to reduce turn-on time of the plurality of transistors.

Abstract: A processing system configured to detect an input object proximate the processing system. The processing system includes sensor circuitry configured to make a determination, when the processing system is in a low ground mass (LGM) state, that a large object is proximate to sensor electrodes of the processing system. The sensor circuitry is further configured, in response to a determination that a large object is proximate the sensor electrodes while the processing system is in the LGM state, to drive a first group of sensor electrodes with one of an inverted signal or a non-inverted signal and drive a second group of sensor electrodes with a static DC voltage.

Abstract: In general, in one aspect, one or more embodiments relate to an input device that includes a proximity sensing panel including sensor electrodes, and a proximity sensing circuit. The proximity sensing circuit is configured to determine that an image refresh rate to an equal or lower frequency than a beacon signal rate. The input-display device is configured to align transmitting a beacon signal on proximity sensing panel to during non-refresh period of a display or perform an additional image refresh frame on the display panel immediately before a next Vsync signal and a corresponding non-refresh period.

Abstract: A classification training system comprises a neural network configured to perform classification of input data, a training dataset including pre-segmented, labeled training samples, and a classification training module configured to train the neural network using the training dataset. The classification training module includes a forward pass processing module, and a backward pass processing module. The backward pass processing module is configured to determine whether a current frame is in a region of target (ROT), determine ROT information such as beginning and length of the ROT and update weights and biases using a cross-entropy cost function and a tunable many-or-one detection (MOOD) cost function, that comprises a tunable hyperparameter for tuning the classifier for a particular task. The backward pass module further computes a soft target value using ROT information and computes a signal output error using the soft target value and network output value.

Abstract: A display driver includes a GRAM, a data driver, and a control circuit. The data driver is configured to: update, in a first mode, display elements of a display panel based on a command provided to the display driver asynchronously with a display vertical sync signal; update, in a second mode, the display elements based on image data stored in the GRAM in synchronization with the display vertical sync signal; and update, in a third mode, the display elements in synchronization with an external vertical sync signal. The control circuit is configured to: switch the display drive to a second mode in response to a first command; adjust, in the second mode, the display vertical sync signal based on an external vertical sync signal; and switch the display driver to the third mode after achieving synchronization of the display vertical sync signal with the external vertical sync signal.

Abstract: A processing system includes sensor circuitry and processing circuitry. The sensor circuitry is configured to, using the sensor electrodes, obtain capacitive measurements of a sensing region, and obtain a resistance measurement of the sensing region. The processing circuitry is coupled to the sensor circuitry. The processing circuitry is configured to determine a location of an input object using the capacitive measurements of the sensing region and determine a force value based on the resistance measurement and the location of the input object. Determining the force value mitigates a temperature variation of the sensing region affecting the resistance measurement. The processing circuitry is further configured to report the force value.

Abstract: Driving a display of an input-display device includes generating, during a first display frame, a first touch sensing waveform to be applied to a touch screen of the display, and generating, during a second display frame, a second touch sensing waveform to be applied to the touch screen. The first touch sensing waveform generates on the display a first touch-to-display noise pattern of touch-to-display noise artifacts. The second touch sensing waveform generates on the display a second touch-to-display noise pattern of touch-to-display noise artifacts. The second touch-to-display noise pattern mitigates the first touch-to-display noise pattern.

Abstract: An input device that includes a processing system, a display panel and an interface is provided. The interface has a base portion with a plurality of base contacts disposed proximal to the display panel and in communication with the processing system, and one or more base reference contacts disposed proximal to the display panel and coupled to a reference signal. The interface also has the rotatable interface with a rotatable contact and an interface reference contact coupled to the rotatable contact and to the one or more base reference contacts. The rotatable interface is configured to rotate about an axis. Rotation of the rotatable interface corresponds to rotation of the rotatable contact and produces a variation of signal strength of resulting signals from at least one of the base contacts. The processing system is configured to determine an angle of rotation based on the variation of signal strength.

Abstract: A secure processing system includes a memory having a secure partition and a non-secure partition, a neural network processing unit (NPU) configured to initiate transactions with the memory, and a memory protection unit (MPU) configured to filter the transactions. Each of the transactions includes at least an address of the memory to be accessed, one of a plurality of first master identifiers (IDs) associated with the NPU, and security information indicating whether the NPU is in a secure state or a non-secure state when the transaction is initiated. The MPU is to selectively deny access to the secure partition of the memory based at least in part on the memory address, the first master ID, and the security information associated with each of the transactions.

Abstract: Systems and method for streaming video content include downscaling video content using a downscaling model to generate downscaled video content and downloading the downscaled video content as a video stream and corresponding upscaling model to a client device. The system converts received video frames to a video memory format comprising channels having the same memory allocation size, each subsequent channel arranged in an adjacent memory location, for input to the downscaling model. The client device upscales the video stream using the received upscaling model for display by the client device in real-time. A training system trains the downscaling model to generate the downscaled video content, based on associated metadata identifying a type of video content. The downscaled video content and associated upscaling models are stored for access by an edge server, which downloads upscaling models to a client device to select an upscaling model.

Abstract: Image compression circuitry comprises first-stage compression circuitry, first-stage selector circuitry, second-stage compression circuitry, and second-stage selector circuitry. The first-stage compression circuitry is configured to sequentially receive a plurality of input blocks each comprising pixel data of a plurality of pixels, generate a plurality of first-stage compressed blocks by compressing the plurality of input blocks, and generate a plurality of first-stage decompressed blocks. The first-stage selector circuitry is configured to select first-stage-selected decompressed blocks from among the plurality of first-stage decompressed blocks and select first-stage-selected compressed blocks corresponding to the first-stage-selected decompressed blocks from among the plurality of first-stage compressed blocks.

Abstract: A biometric sensor may comprise a plurality of a first type of signal traces formed on a first surface of a first layer of a multi-layer laminate package; at least one trace of a second type, formed on a second surface of the first layer or on a first surface of a second layer of the multi-layer laminate package; and connection vias in at least the first layer electrically connecting the signal traces of the first type or the signal traces of the second type to respective circuitry of the respective first or second type contained in an integrated circuit physically and electrically connected to one of the first layer, the second layer or a third layer of the multi-layer laminate package.

Abstract: A display driver comprises an interface and signal supply circuitry. The interface is configured to receive image data. The signal supply circuitry is configured to supply at least one drive control signal to a display panel based on a detection of a data error in the image data associated with a first horizontal line in a first vertical sync period, causing a first pixel circuit and a second pixel circuit hold, in the first vertical sync period, first hold voltages based in part on at least one drive control signal and second hold voltages held in a second vertical sync period prior to the first vertical sync period. The first pixel circuit is associated with the first horizontal line, and the second pixel circuit is associated with a second horizontal line and driven after the first pixel circuit.

Abstract: A display driver comprises image processing circuitry and driver circuitry. The image processing circuitry is configured to receive spatial distribution information of a physical quantity related to a display panel. The image processing circuitry is further configured to generate output voltage data by processing input pixel data associated with subpixels of the pixel. The drive circuitry is configured to drive the display panel based on the output voltage data.

Abstract: An audio spotting system configured for various operating modes including a regular mode and sensitivity mode is described. An example cascade audio spotting system may include a high-power subsystem including a high-power trigger and a transfer module. This high-power trigger includes one or more detection models used to detect whether a target sound activity is included in the one or more audio streams. The one or more detection models are associated with a first set of hyperparameters when the cascade audio spotting system is in a regular mode, and the one or more detection models are associated with a second set of hyperparameters when the cascade audio spotting system is in a sensitivity mode. The transfer module provides at least one of one or more processed audio streams for further processing in response to the high-power trigger detecting the target sound activity in the one or more audio streams.

Abstract: An input-display device includes a display panel, sensor electrodes, and a display driver. The display panel includes source lines. The sensor electrodes are capacitively coupled to the source lines. The display driver is configured to receive image data. The display driver is further configured to process the image data in response to a detection of a horizontal stripe pattern in an image corresponding to the image data. The display driver is further configured to drive the source lines based at least in part on the processed image data.

Abstract: A capacitive sensing layer of an input device includes a substantially rectangular sensor pattern extending in a horizontal direction and in a vertical direction, including first and second electrodes. Each of the first electrodes includes a strip extending in the vertical direction. The first electrodes are separated in the horizontal direction by first gaps. Each of the second electrodes includes a substantially rectangular pad, disposed adjacent to one of the first electrodes, in one of the first gaps. Two of the second electrodes are disposed in each of the first gaps in the horizontal direction. The two of the second electrodes in each of the first gaps are separated by second gaps. A subset of the second electrodes are disposed in each of the first gaps, in the vertical direction. The capacitive sensing layers also includes routing traces of the second electrodes disposed in the second gaps.