Abstract: This disclosure relates generally to image processing and specifically to processing panorama images using neural networks to generate depth maps, layouts, semantic maps or the like with reduced distortion and improved continuity. Methods and systems are described for generating such maps by leveraging several essential properties of these panorama images and by using a panorama panel representation and a neural network framework. A panel geometry embedding network is incorporated for encoding both the local and global geometric features of the panels in order to reduce negative impact of panoramic distortion. A local-to-global transformer network is also incorporated for capturing geometric context and aggregating local information within a panel and panel-wise global context.

Abstract: A method, computer program, and computer system is provided for 3D pose graph reconstruction. 3D reconstruction of indoor buildings for VR/AR applications, such as virtual touring, digital museum, and virtual home sale, may be performed based on estimating pose graph and transformation matrix between cameras by recognizing markers on the ground in panorama images. Image data corresponding to one or more views of a first camera is received. One or more markers corresponding to a second camera are identified within the received image data. A pose graph corresponding to the one or more views of the first camera is constructed based on the identified one or more markers, the pose graph comprising at least an edge.

Abstract: A plurality of two-dimensional (2D) images of the scene is received. Geometric information and semantic information of each of the plurality of 2D images is determined. The geometric information indicates a detected line and a reference direction in the respective 2D image. The semantic information includes classification information of pixels in the respective 2D image. A layout estimation associated with the respective 2D image of the scene is determined based on the geometric information and the semantic information of the respective 2D image. A combined layout estimation associated with the scene is determined based on a plurality of the determined layout estimations associated with the plurality of 2D images of the scene. The Manhattan layout associated with the scene is generated based on the combined layout estimation. The Manhattan layout includes at least a three-dimensional (3D) shape of the scene that includes wall faces orthogonal with respect to each other.

Abstract: Aspects of the disclosure include methods, apparatuses, and non-transitory computer-readable storage mediums for generating a three-dimensional (3D) geometry of a room from a panorama image of the room. An apparatus includes processing circuitry that determine two-dimensional (2D) positions of wall corner points of the room in the panorama image based on a user input. Each wall corner point is in one of a floor plane or a ceiling plane of the room. The processing circuitry calculates 3D positions of the wall corner points based on the 2D positions of the wall corner points, a size of the panorama image, and a distance between the floor plane and a capture position of a device capturing the panorama image, determines a room layout based on an order of the wall corner points, and generates the 3D geometry based on the room layout and the 3D positions of the wall corner points.

Abstract: Aspects of the disclosure include methods, apparatuses, and non-transitory computer-readable storage mediums for generating a floor plan from a point cloud model. An apparatus includes processing circuitry that receives an input three-dimensional point cloud corresponding to a three-dimensional space. The processing circuitry determines a plurality of wall planes in the received input three-dimensional point cloud. The processing circuitry generates a plurality of line segments. Each line segment is generated by projecting a respective wall plane of the plurality of wall planes to a floor plane in the three-dimensional space. The processing circuitry represents the plurality of wall planes in the three-dimensional space using the plurality of line segments in a two-dimensional space corresponding to the floor plan. The processing circuitry adjusts the plurality of line segments in the two-dimensional space to improve the floor plan.

Abstract: Aspects of the disclosure include methods, apparatuses, and non-transitory computer-readable storage mediums for generating a three-dimensional (3D) geometry of a room from a panorama image of the room. An apparatus includes processing circuitry that determine two-dimensional (2D) positions of wall corner points of the room in the panorama image based on a user input. Each wall corner point is in one of a floor plane or a ceiling plane of the room. The processing circuitry calculates 3D positions of the wall corner points based on the 2D positions of the wall corner points, a size of the panorama image, and a distance between the floor plane and a capture position of a device capturing the panorama image, determines a room layout based on an order of the wall corner points, and generates the 3D geometry based on the room layout and the 3D positions of the wall corner points.

Abstract: Aspects of the disclosure include methods, apparatuses, and non-transitory computer-readable storage mediums for generating a floor plan from a point cloud model. An apparatus includes processing circuitry that receives an input three-dimensional point cloud corresponding to a three-dimensional space. The processing circuitry determines a plurality of wall planes in the received input three-dimensional point cloud. The processing circuitry generates a plurality of line segments. Each line segment is generated by projecting a respective wall plane of the plurality of wall planes to a floor plane in the three-dimensional space. The processing circuitry represents the plurality of wall planes in the three-dimensional space using the plurality of line segments in a two-dimensional space corresponding to the floor plan. The processing circuitry adjusts the plurality of line segments in the two-dimensional space to improve the floor plan.

Abstract: A method, computer program, and computer system is provided for 3D pose graph reconstruction. 3D reconstruction of indoor buildings for VR/AR applications, such as virtual touring, digital museum, and virtual home sale, may be performed based on estimating pose graph and transformation matrix between cameras by recognizing markers on the ground in panorama images. Image data corresponding to one or more views of a first camera is received. One or more markers corresponding to a second camera are identified within the received image data. A pose graph corresponding to the one or more views of the first camera is constructed based on the identified one or more markers, the pose graph comprising at least an edge.

Abstract: There is provided an appearance search system comprising one or more cameras configured to capture video of a scene, the video having images of objects. The system comprises one or more processors and memory comprising computer program code stored on the memory and configured when executed by the one or more processors to cause the one or more processors to perform a method. The method comprises identifying one or more of the objects within the images of the objects. The method further comprises implementing a learning machine configured to generate signatures of the identified objects and generate a signature of an object of interest. The system further comprises a network configured to send the images of the objects from the camera to the one or more processors.

Abstract: A method for treating a water-containing substance, including: mixing a water-containing substance with a solvent at a first temperature to obtain a mixture containing a solid substance and a liquid substance; conducting a first separation treatment of the mixture to obtain the solid substance and the liquid substance; and conducting a second separation treatment of the liquid substance obtained by the first separation treatment at a second temperature to obtain an aqueous phase and an organic phase, the organic phase contains the solvent. The mutual solubility of the solvent and water at the first temperature is higher than that of the solvent and water at the second temperature, so that the separation and recovery of water is achieved by liquid-liquid separation without phase transition in the method for treating a water-containing substance, so the energy consumption can be effectively reduced. The invention further relates to a corresponding device.

Abstract: There are described methods and systems for facilitating identification of an object-of-interest. A face similarity score and a body similarity score of a query image relative to a gallery image are determined. A fused similarity score of the query image relative to the gallery image is determined by applying a relationship between the face similarity score, the body similarity score, and the fused similarity score. The fused similarity score is indicative of whether or not the object-of-interest and the potential object-of-interest are the same object-of-interest. For example, a machine learning process is used to fuse the face similarity score and the body similarity into the fused similarity score. The process is repeated for multiple gallery images. The gallery images may then be ranked according to their respective fused similarity scores.

Abstract: There is provided an appearance search system comprising one or more cameras configured to capture video of a scene, the video having images of objects. The system comprises one or more processors and memory comprising computer program code stored on the memory and configured when executed by the one or more processors to cause the one or more processors to perform a method. The method comprises identifying one or more of the objects within the images of the objects. The method further comprises implementing a learning machine configured to generate signatures of the identified objects and generate a signature of an object of interest. The system further comprises a network configured to send the images of the objects from the camera to the one or more processors.

Abstract: There is provided an appearance search system comprising one or more cameras configured to capture video of a scene, the video having images of objects. The system comprises one or more processors and memory comprising computer program code stored on the memory and configured when executed by the one or more processors to cause the one or more processors to perform a method. The method comprises identifying one or more of the objects within the images of the objects. The method further comprises implementing a learning machine configured to generate signatures of the identified objects and generate a signature of an object of interest. The system further comprises a network configured to send the images of the objects from the camera to the one or more processors.

Abstract: Disclosed is an information push method and an electronic device. The method includes: acquiring a page view number of each piece of information in a predetermined information list for a predetermined period of time before a current time; selecting information from the predetermined information list according to the page view number, wherein the selected information is information not meeting a predetermined information displaying condition; acquiring an information recommendation degree of the selected information; determining information to be pushed from the selected information according to the information recommendation degree; and pushing the information to be pushed.

Abstract: Methods, systems, and techniques for sharing layers between convolutional neural networks (CNNs). A data processing system may include a first and a second CNN that share a first group of layers. The first CNN may include the first group of layers in series with a second group of layers and be configured such that data for the first CNN is input to the first group of layers. The second CNN may include the first group of layers in series with a third group of layers and be configured such that data for the second CNN is also input to the first group of layers.

Abstract: Methods, systems, and techniques for classifying an object-of-interest using an artificial neural network, such as a convolutional neural network. An artificial neural network receives a sample image including the object-of-interest overlaying a background and a sample background image excluding the object-of-interest and corresponding to the background overlaid by the object-of-interest. The object-of-interest is classified using the artificial neural network. The artificial neural network classifies the object-of-interest using the sample background and sample images. Prior to receiving the sample background and sample images the artificial neural network has been trained to classify the object-of-interest using training image pairs. Each of at least some of the training image pairs includes a first training image that includes a training object-of-interest overlaying a training background and a training background image excluding the training object-of-interest and corresponding to the training background.

Abstract: Methods, systems, and techniques for classifying an object-of-interest using an artificial neural network, such as a convolutional neural network. An artificial neural network receives a sample image including the object-of-interest overlaying a background and a sample background image excluding the object-of-interest and corresponding to the background overlaid by the object-of-interest. The object-of-interest is classified using the artificial neural network. The artificial neural network classifies the object-of-interest using the sample background and sample images. Prior to receiving the sample background and sample images the artificial neural network has been trained to classify the object-of-interest using training image pairs. Each of at least some of the training image pairs includes a first training image that includes a training object-of-interest overlaying a training background and a training background image excluding the training object-of-interest and corresponding to the training background.

Abstract: There are described methods and systems for facilitating identification of an object-of-interest. A face similarity score and a body similarity score of a query image relative to a gallery image are determined. A fused similarity score of the query image relative to the gallery image is determined by applying a relationship between the face similarity score, the body similarity score, and the fused similarity score. The fused similarity score is indicative of whether or not the object-of-interest and the potential object-of-interest are the same object-of-interest. For example, a machine learning process is used to fuse the face similarity score and the body similarity into the fused similarity score. The process is repeated for multiple gallery images. The gallery images may then be ranked according to their respective fused similarity scores.

Abstract: Disclosed is an information push method and an electronic device. The method includes: acquiring a page view number of each piece of information in a predetermined information list for a predetermined period of time before a current time; selecting information from the predetermined information list according to the page view number, wherein the selected information is information not meeting a predetermined information displaying condition; acquiring an information recommendation degree of the selected information; determining information to be pushed from the selected information according to the information recommendation degree; and pushing the information to be pushed.

Abstract: There is provided an appearance search system comprising one or more cameras configured to capture video of a scene, the video having images of objects. The system comprises one or more processors and memory comprising computer program code stored on the memory and configured when executed by the one or more processors to cause the one or more processors to perform a method. The method comprises identifying one or more of the objects within the images of the objects. The method further comprises implementing a learning machine configured to generate signatures of the identified objects and generate a signature of an object of interest. The system further comprises a network configured to send the images of the objects from the camera to the one or more processors.