Abstract: In some examples, a moisture-proof mat is provided for an intelligent cleaning apparatus or an intelligent cleaning system. The intelligent cleaning apparatus is provided with a wet cleaning pad. The intelligent cleaning apparatus can be charged by a charging station. The moisture-proof mat may include an installation unit, configured to install the moisture-proof mat to the charging station. At least a portion of the moisture-proof mat is configured to be located under the wet cleaning pad (4) when the intelligent cleaning apparatus is charged by the charging station. When the intelligent cleaning apparatus is charged by the charging station, at least a portion of the moisture-proof mat is positioned under the intelligent cleaning apparatus.

Abstract: Embodiments described herein relate generally to electrochemical cells and electrodes with carbon-containing coatings. In some embodiments, an electrochemical cell can include an anode disposed on an anode current collector, a cathode disposed on a cathode current collector, and a separator disposed between the anode and the cathode. The separator has a first side adjacent to the cathode and a second side adjacent to the anode. The electrochemical cell further includes a coating layer disposed on the separator. The coating layer reduces dendrite formation in the electrochemical cell. In some embodiments, the coating layer can include hard carbon. In some embodiments, the coating layer can have a thickness between about 100 nm and about 20 ?m. In some embodiments, the coating layer can be disposed on the first side of the separator.

Abstract: A fireplace with a smart glass panel and use of a smart glass panel in a fireplace are provided. The fireplace comprises a heating mechanism, and a smart glass panel positioned proximate to the heating mechanism. The smart glass panel comprising a transparent pane, where at least a portion of the smart glass panel has optic or thermal transmission properties that may be altered between a first state and a second state.

Abstract: An imaging photoplethysmography (iPPG) system is provided. The iPPG system receives a sequence of images of different regions of the skin of the person, where each region including pixels of different intensities indicative of variation of coloration of the skin. The iPPG system further transforms the sequence of images into a multidimensional time-series signal, each dimension corresponding to a different region from the different regions of the skin. The iPPG system further processes the multidimensional time-series signal with a time-series U-Net neural network wherein the pass-through layers include a recurrent neural network (RNN) to generate a PPG waveform, where the vital sign of the person is estimated based on the PPG waveform, and the iPPG system further renders the estimated vital sign of the person.

Abstract: A computer includes a processor and a memory storing instructions executable by the processor to receive radar data including a radar pixel having a radial velocity from a radar; receive camera data including an image frame including camera pixels from a camera; map the radar pixel to the image frame; generate a region of the image frame surrounding the radar pixel; determine association scores for the respective camera pixels in the region; select a first camera pixel of the camera pixels from the region, the first camera pixel having a greatest association score of the association scores; and calculate a full velocity of the radar pixel using the radial velocity of the radar pixel and a first optical flow at the first camera pixel. The association scores indicate a likelihood that the respective camera pixels correspond to a same point in an environment as the radar pixel.

Abstract: A computer includes a processor and a memory storing instructions executable by the processor to receive radar data from a radar, the radar data including radar pixels having respective measured depths; receive camera data from a camera, the camera data including an image frame including camera pixels; map the radar pixels to the image frame; generate respective regions of the image frame surrounding the respective radar pixels; for each region, determine confidence scores for the respective camera pixels in that region; output a depth map of projected depths for the respective camera pixels based on the confidence scores; and operate a vehicle including the radar and the camera based on the depth map. The confidence scores indicate confidence in applying the measured depth of the radar pixel for that region to the respective camera pixels.

Abstract: An end-to-end trainable grouped mathematically differentiable non-maximal suppression (NMS) technique is presented for monocular 3D object detection. First, formulate NMS as a matrix operation and then group and mask the boxes in an unsupervised manner to obtain a simple closed-form expression of the NMS. This technique addresses the mismatch between training and inference pipelines and, therefore, forces the network to select the best 3D box in a differentiable manner. As a result, the proposed technique achieves state-of-the-art monocular 3D object detection results on the KITTI benchmark dataset performing comparably to monocular video-based methods.

Abstract: The present disclosure relates to a cleaning assembly and a smart cleaning device. The cleaning assembly is capable of being mounted on a machine body of a smart cleaning device and includes a connecting plate and a soft cleaning body, wherein the connecting plate and the soft cleaning body are detachably connected through a slide assembly, the slide assembly includes a mounting groove and a guide bar which are mutually matched, the mounting groove is provided on one of the connecting plate and the soft cleaning body, and the guide bar is disposed on the other one of the connecting plate and the soft cleaning body.

Abstract: An intelligent cleaning device, comprising: a machine body (100); and a cleaning assembly (200) that is detachably connected to the machine body (100) along the forward or backward direction of the machine body (100). The cleaning assembly (200) comprises a liquid containing box, a connecting plate and a flexible cleaning body that are stacked. The liquid containing box is detachably connected to the machine body (100) along the forward or backward direction of the machine body (100), and the liquid containing box at least partially surrounds the outer side of a fan in the machine body (100). The connecting plate is detachably connected to the liquid containing box along the forward or backward direction of the liquid containing box. The flexible cleaning body is detachably connected onto the connecting plate.

Abstract: The present disclosure relates to a sealing structure and a smart cleaning device. The sealing structure is configured to seal an infrared wall-following module of the smart cleaning device. The sealing structure includes a housing and a lens. The housing is connected to a side of the infrared wall-following module facing a protective shell of the smart cleaning device. The lens is connected to the housing, and the lens covers a side surface of the housing facing the protective shell. The lens includes an arc-shaped panel and a connecting plate that is arranged on an edge of the arc-shaped panel and that extends toward the infrared wall-following module.

Abstract: The present disclosure relates to a pneumatic serial device for spraying materials into a deep soil layer and an implementation method thereof. The device comprises a material spraying injector (1), a air blasting system (2) connected with the material spraying injector (1), and a material feeding system (3), wherein the air blasting system (2) and the material feeding system (3) are connected with the material spraying injector (1) in series in sequence, wherein the air blasting system (2) comprises a high-pressure air tank (2-1) and a quick relieve valve (2-2), the material feeding system (3) comprises a material chamber (3-1), a feeding valve (3-2), a valve I (3-3), a material sending chamber (3-4), a valve II (3-5) and a material delivery pipe (3-6) which are connected to each other in sequence, and the material delivery pipe (3-6) extends into the high-pressure air tank (2-1) and is communicated with the quick relieve valve (2-2) at the bottom thereof.

Abstract: Disclosed is an in-situ detection robot for loess geological information, including a housing, an active tensioning system, an auxiliary tensioning system, a control system, a drive system, and a detection system. The active tensioning system is configured to provide the robot with a tensioning force in an aperture direction of a loess hole for the robot to travel in the loess hole. The auxiliary tensioning system is configured to provide the robot with a pre-tensioning force for the robot to travel in the loess hole. The control system is configured to control a movement mode of the robot according to different travelling environments of the robot in the loess hole. The driving system is configured to provide the robot with a travelling power in the loess hole. The detection system is configured to detect related parameters of the loess geological information.

Abstract: Gait, the walking pattern of individuals, is one of the important biometrics modalities. Most of the existing gait recognition methods take silhouettes or articulated body models as gait features. These methods suffer from degraded recognition performance when handling confounding variables, such as clothing, carrying and viewing angle. To remedy this issue, this disclosure proposes to explicitly disentangle appearance, canonical and pose features from RGB imagery. A long short-term memory integrates pose features over time as a dynamic gait feature while canonical features are averaged as a static gait feature. Both of them are utilized as classification features.

Abstract: Gait, the walking pattern of individuals, is one of the most important biometrics modalities. Most of the existing gait recognition methods take silhouettes or articulated body models as the gait features. These methods suffer from degraded recognition performance when handling confounding variables, such as clothing, carrying and view angle. To remedy this issue, a novel AutoEncoder framework is presented to explicitly disentangle pose and appearance features from RGB imagery and a long short-term memory integration of pose features over time produces the gait feature.

Abstract: Apparatus, systems, and methods described herein relate to the manufacture and use of electrochemical cells with a flame retardant mechanism. In some embodiments, an electrochemical cell includes a first current collector coupled to a first portion of a first pouch, the first current collector having a first electrode material disposed thereon. The electrochemical cell further includes a second current collector coupled to a second portion of the first pouch, the second current collector having a second electrode material disposed thereon. The electrochemical cell further includes a separator disposed between the first electrode material and the second electrode material, the first portion of the first pouch coupled to the second portion of the first pouch to enclose the electrochemical cell. The electrochemical cell further includes a flame retardant material coated to the first pouch and a second pouch, the second pouch enclosing the first pouch and the flame retardant material.

Abstract: A system and method are provided for tracking magnetically-labeled substances, such as transplanted cells, in subjects using magnetic resonance imaging (MRI).

Abstract: A controller for executing a task based on probabilistic image-based landmark localization, uses a neural network, which is trained to process images of objects of a type having a structured set of landmarks to produce a parametric probability distribution defined by values of parameters for a location of each landmark in each processed image. The controller submits the set of input images to the neural network to produce the values of the parameters that define the parametric probability distribution over the location of each landmark in the structured set of landmarks of each input image. Further, the controller determines, for each input image, a global landmark uncertainty for the image based on the parametric probability distributions of landmarks in the input image and executes the task based on the parametric probability distributions of landmarks in each input image and the global landmark uncertainty of each input image.

Abstract: The present disclosure provides a method, an apparatus, and a medium for calculating capacities of a plurality of photovoltaic power stations. The method includes: receiving a plurality of historical weather data sets from the plurality of weather monitoring stations; determining a scene year based on the plurality of historical weather data sets; receiving an actual generating capacity of each photovoltaic power station predicted by a power system dispatch center; extracting weather data of each photovoltaic power station in the scene year from the corresponding historical weather data set; obtaining an available generating capacity of each photovoltaic power station in the scene year based on the weather data; and determining a capacity of each photovoltaic power station in the scene year according to the actual generating capacity and the available generating capacity of each photovoltaic power station.