Abstract: An intelligent switch device and a power generation system are provided. The intelligent switch device includes a first power port, a second power port, a switch unit and a control unit. A first end of a switch unit is connected to a first power supply via a first power port, and a second end of the switch unit is connected to a second power supply via a second power port. The control unit is configured to control the switch unit to be turned on for a first time period when the first power supply is in abnormal condition, where the abnormal condition causes the switch unit to be tripped. The first time period is greater than or equal to a time period required for a device to respond to the abnormal condition.

Abstract: Systems and methods for text segmentation are described. Embodiments of the inventive concept are configured to receive an image including a foreground text portion and a background portion, classify each pixel of the image as foreground text or background using a neural network that refines a segmentation prediction using a key vector representing features of the foreground text portion, wherein the key vector is based on the segmentation prediction, and identify the foreground text portion based on the classification.

Abstract: Systems and methods for texture hallucination with a large upscaling factor are described. Embodiments of the systems and methods may receive an input image and a reference image, extract an upscaled feature map from the input image, match the input image to a portion of the reference image, wherein a resolution of the reference image is higher than a resolution of the input image, concatenate the upscaled feature map with a reference feature map corresponding to the portion of the reference image to produce a concatenated feature map, and generate a reconstructed image based on the concatenated feature map using a machine learning model trained with a texture loss and a degradation loss, wherein the texture loss is based on a high frequency band filter, and the degradation loss is based on a downscaled version of the reconstructed image.

Abstract: The present invention discloses a core assembly, a sheath assembly, a sheath, a sheath processing method and an interventional instrument delivery system and method. A core assembly for delivering an interventional instrument according to the present invention comprises a core tube, a locking member fixed at a distal end of the core tube for connecting the interventional instrument, and a bendable adjustable tube mounted around an outer periphery of the core tube, wherein the distal ends of the bendable adjustable tube and the core tube are fixedly connected to each other, and the proximal ends can slide relative to each other.

Abstract: A method for capacity expansion of an energy storage system and an energy storage system are provided. The method is applied to a controller in the energy storage system. The controller acquires a value of a characteristic parameter of a to-be-added second energy storage device as a target characteristic value. Then, the controller controls a charge/discharge power converter to charge or discharge first energy storage devices until a proximity of a value of a characteristic parameter of at least one first energy storage device to the target characteristic value is less than a preset proximity. Finally, a signal indicating that the second energy storage device is allowed to be connected into the energy storage system is generated and outputted to notify an operator.

Abstract: Systems and methods for texture hallucination with a large upscaling factor are described. Embodiments of the systems and methods may receive an input image and a reference image, extract an upscaled feature map from the input image, match the input image to a portion of the reference image, wherein a resolution of the reference image is higher than a resolution of the input image, concatenate the upscaled feature map with a reference feature map corresponding to the portion of the reference image to produce a concatenated feature map, and generate a reconstructed image based on the concatenated feature map using a machine learning model trained with a texture loss and a degradation loss, wherein the texture loss is based on a high frequency band filter, and the degradation loss is based on a downscaled version of the reconstructed image.

Abstract: The present disclosure relates to systems, computer-implemented methods, and non-transitory computer readable medium for automatically transferring makeup from a reference face image to a target face image using a neural network trained using semi-supervised learning. For example, the disclosed systems can receive, at a neural network, a target face image and a reference face image, where the target face image is selected by a user via a graphical user interface (GUI) and the reference face image has makeup. The systems transfer, by the neural network, the makeup from the reference face image to the target face image, where the neural network is trained to transfer the makeup from the reference face image to the target face image using semi-supervised learning. The systems output for display the makeup on the target face image.

Abstract: Systems and methods provide for generating glyph initiations using a generative font system. A glyph variant may be generated based on an input vector glyph. A plurality of line segments may be approximated using a differentiable rasterizer with the plurality of line segments representing the contours of the glyph variant. A bitmap of the glyph variant may then be generated based on the line segments. The image loss between the bitmap and a rasterized representation of a vector glyph may be calculated and provided to the generative font system. Based on the image loss, a refined glyph variant may be provided to a user.

Abstract: Provided are a network operation support system and a network device management method, belonging to the technical field of communications. The system uses a micro-service architecture realized based on a container, and the system itself has a good expansibility and availability, and supports automation and layout in a cloud environment. In addition, the system also uses a main database to synchronously store operating data of various micro-services for unified maintenance of the data, thereby improving the analysis efficiency of the system.

Abstract: A stent and a securely-installed artificial valve replacement device having the same, the stent being of a cylindrical structure; the top of the stent is provided with a fixed ear (60); the fixed ear (60) has a neck portion (601) connected to the top of the stent, and a head portion (602) engaged with the fixed head of the stent; the head portion (602) has a bending structure for improving the overall radial thickness; and the artificial valve replacement device is comprised of a stent and a prosthetic valve fixed on the stent. The stent with a bending structure overcomes the problem of easily disengaging from the fixed head of the stent, while not affecting the release of the stent.

Abstract: A bendable sheath and a delivery system using the bendable sheath. The bendable sheath comprises a tube body (3). The tube body (3) comprises a distal end and a proximal end. A tube wall of the tube body (3) is connected to a pull wire (8). One end of the pull wire (8) extends towards the proximal end of the tube body (3), and the other end is connected to the tube body (3) near the distal end of the tube body (3). The pull wire (8) comprises at least a section thereof disposed freely outside the tube body (3) and near the distal end of the tube body (3). The pull wire (8) in the bendable sheath comprises the section disposed freely outside the sheath tube body (3) and, when pulled, the section is disposed so as to facilitate the application of force. The section moves relative to the tube body (3), such that a force application point is adaptively changed.

Abstract: In implementations of super-resolution with reference images, a super-resolution image is generated based on reference images. Reference images are not constrained to have same or similar content as a low-resolution image being super-resolved. Texture features indicating high-frequency content are extracted into texture feature maps, and patches of texture feature maps of reference images are determined based on texture feature similarity. A content feature map indicating low-frequency content of an image is adaptively fused with a swapped texture feature map including patches of reference images with a neural network based on similarity of texture features. A user interfaces allows a user to select regions of multiple reference images to use for super-resolution. Hence, a super-resolution image can be generated with rich texture details incorporated from multiple reference images, even in the absence of reference images having similar content to an image being upscaled.

Abstract: A stent and a securely-installed artificial valve replacement device having the same, the stent being of a cylindrical structure; the top of the stent is provided with a fixed ear (60); the fixed ear (60) has a neck portion (601) connected to the top of the stent, and a head portion (602) engaged with the fixed head of the stent; the head portion (602) has a bending structure for improving the overall radial thickness; and the artificial valve replacement device is comprised of a stent and a prosthetic valve fixed on the stent. The stent with a bending structure overcomes the problem of easily disengaging from the fixed head of the stent, while not affecting the release of the stent.

Abstract: An easy-to-control interventional instrument delivery device, comprising a core tube (7), a guiding head (2) and a fixing head (3) being fixed onto the core tube (7); the guiding head (2) is fixed at a distal end of the core tube (7), and the fixing head (3) extends from a proximal end side of the core tube (7), an interventional instrument mounting position being located between the guiding head (2) and the fixing head (3); the outer periphery of the interventional instrument mounting position is provided with an axially slidable outer sheathing tube (5), further being provided with a floating limiting strip (1); a proximal end of the floating limiting strip (1) is a starting end (13) that is connected near the fixed head (3), and a distal end is a terminal end (11, 101) that floats between the interventional instrument mounting position and the outer sheathing tube (5).

Abstract: A thrombus filter has a sheath, a thrombus intercepting net having an intercepting port, and an overfill prevention umbrella being a meshed structure that always surrounds the entire thrombus intercepting net. The overfill prevention umbrella has an umbrella stand and an umbrella surface, and is self-biased such that, at body temperature, the umbrella surface is configured to always converge towards the center of the intercepting port. The thrombus filter assumes a first position when the thrombus intercepting net is released from the sheath, where the distal opening of the overfill prevention umbrella is pushed open by the intercepting port. The thrombus filter assumes a second position when the thrombus intercepting net is retrieved into the sheath, where the umbrella surface converges towards the center of the intercepting port, and the open end of the umbrella surface is constricted to prevent the thrombus from escaping.

Abstract: High resolution style transfer techniques and systems are described that overcome the challenges of transferring high resolution style features from one image to another image, and of the limited availability of training data to perform high resolution style transfer. In an example, a neural network is trained using high resolution style features which are extracted from a style image and are used in conjunction with an input image to apply the style features to the input image to generate a version of the input image transformed using the high resolution style features.

Abstract: The present invention provides an anti-fracture sheath. The anti-fracture sheath comprises a tube body and an expansion section for accommodating an implantable device, wherein one side of the distal end of the tube body is joined with the expansion section, at least two reinforcing ribs are axially arranged on the tube wall of the expansion section in an internally extending manner, and the proximal ends of the reinforcing ribs extend from the expansion section and are fixedly connected with the tube wall of the tube body; and the materials of the reinforcing ribs are Kevlar fibers, the diameter of each reinforcing rib is from 0.03 mm to 0.12 mm, and the number of the reinforcing ribs is from two to four. The anti-fracture sheath provided by the present invention has the advantage that the strength of the expansion section is improved when the whole compliance of the sheath is maintained, thereby avoiding fracture of the sheath in an implantable device releasing process.

Abstract: In implementations of super-resolution with reference images, a super-resolution image is generated based on reference images. Reference images are not constrained to have same or similar content as a low-resolution image being super-resolved. Texture features indicating high-frequency content are extracted into texture feature maps, and patches of texture feature maps of reference images are determined based on texture feature similarity. A content feature map indicating low-frequency content of an image is adaptively fused with a swapped texture feature map including patches of reference images with a neural network based on similarity of texture features. A user interfaces allows a user to select regions of multiple reference images to use for super-resolution. Hence, a super-resolution image can be generated with rich texture details incorporated from multiple reference images, even in the absence of reference images having similar content to an image being upscaled.

Abstract: High resolution style transfer techniques and systems are described that overcome the challenges of transferring high resolution style features from one image to another image, and of the limited availability of training data to perform high resolution style transfer. In an example, a neural network is trained using high resolution style features which are extracted from a style image and are used in conjunction with an input image to apply the style features to the input image to generate a version of the input image transformed using the high resolution style features.

Abstract: Provided are a network operation support system and a network device management method, belonging to the technical field of communications. The system uses a micro-service architecture realized based on a container, and the system itself has a good expansibility and availability, and supports automation and layout in a cloud environment. In addition, the system also uses a main database to synchronously store operating data of various micro-services for unified maintenance of the data, thereby improving the analysis efficiency of the system.