Abstract: A lithium-ion battery electrode includes a current collector and n electrode plate layers laminated on the current collector, n being an integer greater than or equal to 2. The electrode plate layer contains a pore-forming agent. A content of the pore-forming agent in the electrode plate layers gradually increases along a direction gradually away from the current collector. The pore-forming agent is selected from an electrolyte solution additive which is solid at normal temperature.

Abstract: A display substrate includes: a base substrate (100); a plurality of sub-pixels (R, G, B) located on the base substrate (100), every two rows of sub-pixels (R, G, B) constituting a pixel group; a plurality of first gate lines (Gate1) located at first row gaps between the pixel groups, two first gate lines (Gate1) being arranged at each first row gap; and a plurality of photosensors (101), the orthographic projection of each row of photosensors (101) on the base substrate (100) completely covering a second row gap in the pixel group and partially overlapping with the orthographic projections of the sub-pixels (R, G, B), thereby avoiding the bright and dark difference between adjacent rows and ensuring the aperture ratio.

Abstract: Methods and systems for high-resolution image manipulation are disclosed. An original high-resolution image to be manipulated is obtained, as well as a driving signal indicating a manipulation result. The original high-resolution image is down-sampled to obtain a low-resolution image to be manipulated. Using a trained manipulation generator, a low-resolution manipulated image and a motion field are generated from the low-resolution image. The motion field represent pixel displacements of the low-resolution image to obtain the manipulation indicated by the driving signal. A high-frequency residual image is computed from the original high-resolution image. A high-frequency manipulated residual image is generated using the motion field. A high-resolution manipulated image is outputted by combining the high-frequency manipulated residual image and a low-frequency manipulated image generated from the low-resolution manipulated image by up-sampling.

Abstract: Embodiments are disclosed for articulated part extraction using images of animated characters from sprite sheets by a digital design system. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving an input including a plurality of images depicting an animated character in different poses. The disclosed systems and methods further comprise, for each pair of images in the plurality of images, determining, by a first machine learning model, pixel correspondences between pixels of the pair of images, and determining, by a second machine learning model, pixel clusters representing the animated character, each pixel cluster corresponding to a different structural segment of the animated character. The disclosed systems and methods further comprise selecting a subset of clusters that reconstructs the different poses of the animated character. The disclosed systems and methods further comprise creating a rigged animated character based on the selected subset of clusters.

Abstract: Embodiments are disclosed for articulated part extraction using images of animated characters from sprite sheets by a digital design system. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving an input including a plurality of images depicting an animated character in different poses. The disclosed systems and methods further comprise, for each pair of images in the plurality of images, determining, by a first machine learning model, pixel correspondences between pixels of the pair of images, and determining, by a second machine learning model, pixel clusters representing the animated character, each pixel cluster corresponding to a different structural segment of the animated character. The disclosed systems and methods further comprise selecting a subset of clusters that reconstructs the different poses of the animated character. The disclosed systems and methods further comprise creating a rigged animated character based on the selected subset of clusters.

Abstract: A data processing method, apparatus, a device, and a computer-readable medium are provided for use in fields such as artificial intelligence, assisted driving, and the like. An object pose detection result corresponding to an object in an image frame and a part pose detection result corresponding to a first object part of the object in the image frame is acquired. At least one object part of the object is missing from the object pose detection result. The first object part is one or more parts of the object. Interpolation processing is performed on the at least one object part missing from the object pose detection result according to the part pose detection result and a standard pose associated with the object to obtain a global pose corresponding to the object. The global pose is used for controlling a computer to realize a service function corresponding to the global pose.

Abstract: A method for synchronous production of manganese tetraoxide and ferric oxide for a soft magnetic material by using marine polymetallic nodules includes: 1) crushing and grinding marine manganese nodules and baking to a constant weight; thoroughly mixing with a mixed flux and roasting in a muffle furnace; 2) carrying out solid-liquid separation, washing solid-phase precipitates with water, grinding the solid, adding sulfuric acid, controlling the temperature to be below 50° C., and vacuuming a reactor up; 3) adding a reducing agent to react at room temperature for 5-10 min, adding ammonia water to adjust the pH value to 5.5, and carrying out separation and filtering; 4) controlling the temperatures of manganese sulfate and ferric sulfate solutions to be below 50° C., and adding ammonium sulfide; and 5) washing with deionized water, and calcining at 800-900° C. for 1-3 s by a suspension low-temperature instantaneous firing system.

Abstract: Systems, methods, and computer-readable media for identifying a main group of people in an image via social relation recognition. The main group of people is identified within an image by identifying social relationships between people visible in the image. The identification of social relationships is performed by a Social Relation Recognition Network (SRRN) trained using deep learning. The SRRN combines two techniques for group identification, First Glance and Graph Reasoning, and fuses their outputs to generate a prediction of group membership. A group refinement module improves and filters the group membership after identification of an initial main group.

Abstract: Techniques for using multiple voice-enabled devices in a user environment to reduce the latency for obtaining responses to user utterances from a remote system. The voice-enabled devices may each establish connections with the remote system to have the remote system perform supplemental speech processing for utterances the devices are unable to process locally. One voice-enabled device may have a higher-latency connection to the remote system, and another voice-enabled device may have a lower-latency connection to the remote system. The lower-latency device may send an utterance to the remote system before the higher-latency device is able, and the remote system may begin processing the utterance faster than if the lower-latency device sent the utterance. The remote system may then provide a response for the utterance to the higher-latency device in less time than if the remote system had to wait for the utterance from the higher-latency device.

Abstract: Methods and systems for fully-automatic image processing to detect and remove unwanted people from a digital image of a photograph. The system includes the following modules: 1) Deep neural network (DNN)-based module for object segmentation and head pose estimation; 2) classification (or grouping) of wanted versus unwanted people based on information collected in the first module; 3) image inpainting of the unwanted people in the digital image. The classification module can be rules-based in an example. In an example, the DNN-based module generates, from the digital image: 1. A list of object category labels, 2. A list of object scores, 3. A list of binary masks, 4. A list of object bounding boxes, 5. A list of crowd instances, 6. A list of human head bounding boxes, and 7. A list of head poses (e.g., yaws, pitches, and rolls).

Abstract: This disclosure provides for methods and a system for multiple instance segmentation and tracking. According to an aspect a method is provided. The method includes sending an image to a backbone network and generating image feature outputs. The method further includes sending the image feature outputs to a spatial attention module for generating a feature map associated with objects in the image. The method further includes sending the feature map to a category feature module for generating an instance category output indicating the objects. The method further includes sending the image feature outputs to a mask generating module for generating masks. The method further includes generating: the instance category output via the category feature module, and the masks via the mask generating module. In some embodiments, the method further includes generating re-identification embedding information associated with the objects based on image feature outputs.

Abstract: Methods and systems for high-resolution image manipulation are disclosed. An original high-resolution image to be manipulated is obtained, as well as a driving signal indicating a manipulation result. The original high-resolution image is down-sampled to obtain a low-resolution image to be manipulated. Using a trained manipulation generator, a low-resolution manipulated image and a motion field are generated from the low-resolution image. The motion field represent pixel displacements of the low-resolution image to obtain the manipulation indicated by the driving signal. A high-frequency residual image is computed from the original high-resolution image. A high-frequency manipulated residual image is generated using the motion field. A high-resolution manipulated image is outputted by combining the high-frequency manipulated residual image and a low-frequency manipulated image generated from the low-resolution manipulated image by up-sampling.

Abstract: Methods and systems for determining human-object interactions (HOIs) in images are provided. The method includes receiving an image. The method further includes detecting at least one human in the image, and at least one object in the image. The method further includes creating one or more proposals, wherein each proposal includes a human of the at least one human and an object of the at least one object. The method further includes determining whether an HOI exist in each of the one or more proposals. In some embodiments, the method further includes generating a mask for each proposal of the one or more proposals in which an HOI is determined to exist, the mask generated based on the determined HOI. In some embodiment, the HOI determination is based on one or more of extracted human feature information, extracted object appearance feature information, and extracted spatial feature information.

Abstract: Methods and systems for high-resolution image inpainting are disclosed. An original high-resolution image to be inpainted is obtained, as well as an inpainting mask indicating an inside-mask area to be inpainted. The original high-resolution image is down-sampled to obtain a low-resolution image to be inpainted. Using a trained inpainting generator, a low-resolution inpainted image and a set of attention scores are generated from the low-resolution image. The attention scores represent the similarity between inside-mask regions and outside-mask regions. A high-frequency residual image is computed from the original high-resolution image. An aggregated high-frequency residual image is generated using the attention scores, including high-frequency residual information for the inside-mask area. A high-resolution inpainted image is outputted by combining the aggregated high-frequency residual image and a low-frequency inpainted image generated from the low-resolution inpainted image.

Abstract: A crowd motion summarization method that provides a rich, real-time description of the crowd's characteristics from a video, such as, speed, orientation, count, spatial locations, and time. A feature tracking module receives each video frame and detects features (feature points) from the video frame. A crowd occupancy detection module receives the video frame and generates a binary crowd occupancy map having human pixel positions which indicate the human location versus non-human location, and generates a total human count of humans detected in the video frame. The feature tracking module generates feature tracking information for only those features contained in the human pixel positions which indicate the human location. In an example, the detected features are Kanade-Lucas-Tomasi (KLT) features. A feature-crowd matching module generates, using the feature tracking information and the total human count: crowd motion data. The method outputs the crowd motion data.

Abstract: A crowd motion summarization method that provides a rich, real-time description of the crowd's characteristics from a video, such as, speed, orientation, count, spatial locations, and time. A feature tracking module receives each video frame and detects features (feature points) from the video frame. A crowd occupancy detection module receives the video frame and generates a binary crowd occupancy map having human pixel positions which indicate the human location versus non-human location, and generates a total human count of humans detected in the video frame. The feature tracking module generates feature tracking information for only those features contained in the human pixel positions which indicate the human location. In an example, the detected features are Kanade-Lucas-Tomasi (KLT) features. A feature-crowd matching module generates, using the feature tracking information and the total human count: crowd motion data. The method outputs the crowd motion data.

Abstract: Methods and systems for fully-automatic image processing to detect and remove unwanted people from a digital image of a photograph. The system includes the following modules: 1) Deep neural network (DNN)-based module for object segmentation and head pose estimation; 2) classification (or grouping) of wanted versus unwanted people based on information collected in the first module; 3) image inpainting of the unwanted people in the digital image. The classification module can be rules-based in an example. In an example, the DNN-based module generates, from the digital image: 1. A list of object category labels, 2. A list of object scores, 3. A list of binary masks, 4. A list of object bounding boxes, 5. A list of crowd instances, 6. A list of human head bounding boxes, and 7. A list of head poses (e.g., yaws, pitches, and rolls).

Abstract: Methods and systems for high-resolution image inpainting are disclosed. An original high-resolution image to be inpainted is obtained, as well as an inpainting mask indicating an inside-mask area to be inpainted. The original high-resolution image is down-sampled to obtain a low-resolution image to be inpainted. Using a trained inpainting generator, a low-resolution inpainted image and a set of attention scores are generated from the low-resolution image. The attention scores represent the similarity between inside-mask regions and outside-mask regions. A high-frequency residual image is computed from the original high-resolution image. An aggregated high-frequency residual image is generated using the attention scores, including high-frequency residual information for the inside-mask area. A high-resolution inpainted image is outputted by combining the aggregated high-frequency residual image and a low-frequency inpainted image generated from the low-resolution inpainted image.

Abstract: A cell radio network temporary identifier (C-RNTI) allocation method and device are disclosed. The method includes: when UE accesses a serving cell, randomly determining, by an entity at which the serving cell is located, a start position for selecting a C-RNTI, selecting a C-RNTI for the UE within a C-RNTI interval range starting from the start position, and allocating the selected C-RNTI to the UE. So the entity at which the serving cell is located randomizes the start position of allocation performed by the entity and uses the start position as a start position of an allocation interval, so that allocation of the C-RNTI is randomized.

Abstract: An uplink power control method and apparatus are provided. The method is performed by a base station and includes: determining, according to a first quantity of accesses and a second quantity of accesses, whether to adjust a cell-level uplink power parameter, where the first quantity of accesses is a count of times that UEs access a cell of the base station by sending a random access preamble for one time, and the second quantity of accesses is a count of times that the UEs access the cell by sending a random access preamble for at least two times; and adjusting, when determining to adjust the cell-level uplink power parameter, the cell-level uplink power parameter, and sending an adjusted parameter to UE within the coverage area of the cell, so that the UE sends an uplink signal according to the adjusted cell-level uplink power parameter, thereby saving a measurement resource.