Abstract: Disclosed are method and apparatus for fusing an object into a panoramic video. The method may include: playing a panoramic video in a panoramic video player; obtaining a target object; parsing the target object to obtain a coordinates parameter and a material parameter of the target object; setting the color of each pixel of the target object when the target object is fused into the panoramic video according to the coordinate parameter and the material parameter, so as to obtain a colored target object; and fusing the colored target object into the panoramic video.

Abstract: The present disclosure relates to a voice shopping method, device, and a computer-readable storage medium, and relates to the field of voice recognition technology. The method includes: receiving a query instruction from a user by voice, and performing semantic recognition on the query instruction to determine one or more keywords of query content of the user; determining a search range of the keywords according to one or more shopping behavior records of the user; searching the keywords within the search range to obtain commodity information related to the query content, and performing voice output; and receiving an ordering instruction from the user by voice, and performing semantic recognition on the ordering instruction to determine whether an order is made.

Abstract: The present disclosure discloses a method and a device for acquiring figure parameters of a user. The method includes the following steps: acquiring a photo of the user; processing the photo to generate a profile of the user; taking a parameterized three-dimensional human body model with a projection profile consistent with the profile of the user as a target parameterized three-dimensional human body model; and taking figure parameters of the target parameterized three-dimensional human body model as figure parameters of the user.

Abstract: The present disclosure relates to an article conveying system and an article conveying method. The article conveying system comprises: at least two first conveying channels configured to be arranged horizontally on a building; at least one second conveying channel configured to be arranged vertically on the building; a conveyer configured to move on the first conveying channels and the second conveying channel to convey an article to different positions of the building; and transfer device arranged between the first conveying channels and the second conveying channel, to transfer the article and/or the conveyer from the first conveying channels to the second conveying channel or from the second conveying channel to the first conveying channels. The present disclosure can save physical strength and time of delivery persons and consignees, effectively achieve transfer of the article and/or the conveyer between the first conveying channels and the second conveying channel, and improve the conveying efficiency.

Abstract: The present disclosure provides a path planning method and device for an unmanned device. The method includes taking a current landmark as a current node, obtaining a landmark corresponding to a product to be picked with a route closest to the current node, and taking the obtained landmark as the current node, wherein an initial value of the current node is an initial landmark of the unmanned device; determining whether the current node is a landmark corresponding to the last product to be picked; if no, performing the step of obtaining a landmark corresponding to a product to be picked with a route closest to the current node; and if yes, taking a route composed of the initial landmark of the unmanned device and landmarks corresponding to products to be picked that are selected sequentially as a travel route of the unmanned device.

Abstract: The present disclosure relates to the technical field of multimedia, and provides a video processing method and apparatus.

Abstract: A method and a device for extracting a core word of a commodity short text, is provided. The method of extracting a core word of a commodity short text includes: obtaining commodity short texts in a data set; performing word segmenting on each commodity short text; obtaining a document vector of the commodity short text according to context information of word segments of the commodity short text; clustering the commodity short texts in the data set according to document vectors; determining a cluster level weight of each word segment of the commodity short text in a category to which the commodity short text belongs; and determining a core word of the commodity short text according to the cluster level weight of each word segment.

Abstract: The invention provides a call-out method and apparatus for an immediate callback, an electronic device, and a computer readable medium, and can solve the problem that the existing call-out scheme of the preview type cannot effectively implement an immediate callback, and can improve user experience, increase an agent utilization rate, and reduce a pressure on a system caused by frequent scheduling of timed tasks. The call-out method for an immediate callback of the invention comprises: receiving an immediate callback request of a user, the immediate callback request including a callback number and submitted service data; determining, according to pre-stored mapping relationships between service data and agent's telephone skills, agents processing the submitted service data; and judging whether a ready agent exists in the agents, and if so, immediately calling the callback number; otherwise, calling the callback number when there is a ready agent.

Abstract: A commodity recommendation method, apparatus and system, and a computer readable storage medium. The method includes: analyzing an image of a specific offline customer currently entering a store to obtain an attribute of the specific offline customer; determining an online user matching the specific offline customer according to the attribute of the specific offline customer; constructing a collection of popular commodities of the store according to commodities corresponding to stay positions of offline customers who have entered the store before; and recommending a commodity to the specific offline customer according to a historical shopping information of the online user and the collection of popular commodities.

Abstract: The present disclosure relates to a method and apparatus for determining a target rotation direction. Said method for determining the target rotation direction comprises: inputting successive video frames which comprise a rotation target; establishing a background model for the first image frame in said video frames; performing foreground detection on the video frames other than the first frame by means of said background model so as to determine the rotation axis of said rotation target; obtaining the distribution of optical flow points within a preset region of said rotation axis; determining the direction of rotation of said rotation target according to the distribution of optical flow points within said preset region. By means of the present disclosure, it is possible to simply and efficiently determine the clock direction of a rotation target in a video.

Abstract: A video sharing implementation method and system are provided, with the method including receiving a user's access request for a video sharing link which carries an address of a video and an identifier of a shared object, parsing the video sharing link to obtain the address of the video and the identifier of the shared object, playing the video according to the address of the video, and displaying the shared object corresponding to the identifier of the shared object in a video display interface, and jumping to a product link of the shared object in response to the user's selection of the shared object.

Abstract: An interface information display method and apparatus are provided. The method includes: in a list display state, if a click on a list item is detected, performing operations including: acquiring object information and a position of the clicked list item; generating a screenshot image of a current page; displaying the screenshot image in full screen, displaying a first mask layer on the screenshot image, and displaying a second mask layer on the first mask layer; shrinking the screenshot image to a predetermined ratio, acquiring an object thumbnail and detailed information from the object information, displaying, in a predetermined size, the object thumbnail on the second mask layer, and displaying the detailed information in a blank area on the second mask layer; and returning to the list display state when an area of the first mask layer non-overlapping with the second mask layer is clicked.

Abstract: A method and apparatus for delivering an article are provided. The method may include: in response to reception of a vehicle unlocked notification message sent by a terminal device of a user, sending the vehicle unlocked notification message to a terminal device of a deliveryman; in response to reception of an article delivered notification message sent by the terminal device of the deliveryman, sending the article delivered notification message to the terminal device of the user; in response to reception of a vehicle locked notification message sent by the terminal device of the user, sending the vehicle locked notification message to the terminal device of the deliveryman.

Abstract: The disclosure provides an articles picking method, a control system and an articles picking system, and relates to the technical field of intelligent warehousing. The articles picking method of the present disclosure includes: receiving an order of a single articles picking point of an AGV arriving at an articles picking point, the order of the single articles picking point comprising information of articles belonging to a same articles picking point in one or more orders associated with the AGV; recommending the order of the single articles picking point to articles pickers of no less than a predetermined number; setting the order of the single articles picking point to be an assigned state to avoid the order of the single articles picking point being picked repeatedly in the event that there is an articles picker taking the order of the single articles picking point.

Abstract: Provided are a method and apparatus for determining trajectory, and a method, apparatus and system for recommending time. The method includes obtaining trajectory information of multiple trajectory points of a first object, the trajectory information including the geographic location of the trajectory point and the time when the first object is located at the trajectory point; clustering the trajectory points according to the geographic location of each trajectory point to obtain multiple regions not overlapped with each other; determining the time corresponding to each region according to the time of at least one trajectory point in each region; arranging the regions according to the time corresponding to each region to obtain a regional motion trajectory of the first object; and determining a frequent trajectory of the first object based on multiple regional motion trajectories, the frequent trajectory including a frequent region and a corresponding frequent time period.

Abstract: Disclosed in embodiments of the present application are a QR code generation method and apparatus for a terminal device. The terminal device comprises a camera. A specific implementation mode of the method comprises: receiving a QR code generation instruction input by a user, wherein the QR code generation instruction comprises a QR code information; starting a camera, and generating a camera real-time photographing picture; and generating a real-time picture displaying the QR code based on the QR code information and the generated camera real-time photographing picture. The implementation mode enables the user to clearly distinguish whether the QR code is an originally generated QR code, thereby improving the security of the QR code.

Abstract: Disclosed are an image processing method and apparatus. The method may include: inputting a target image into a pre-trained image detection model to obtain a feature value of a rectangular box in which each graphic of at least one graphic presented by the target image is located, and a probability value of each graphic being a specified graphic code; in response to determining at least two rectangular boxes having regions partially overlapped in the rectangular box involved in the at least one graphic, determining a ratio of an area of partially overlapped regions enclosed by the at least two rectangular boxes to the sum of the areas of the regions enclosed by the at least two rectangular boxes; based on the comparison between the ratio and a preset ratio threshold value, determining a target rectangular box from the at least two rectangular boxes; and generating a processed target image.

Abstract: The present disclosure provides a pedestrian re-identification method and apparatus. The pedestrian re-identification apparatus detects a pedestrian to be identified in a preset video frame, extract all pedestrians appearing in video frames shot in a preset time period before a shooting time of the preset video frame as historical pedestrians, calculate a feature distance between the pedestrian to be identified and each historical pedestrian of the historical pedestrians, extract historical pedestrian IDs corresponding to a preset number of the feature distances in ascending order of the feature distances, and cluster the historical pedestrian IDs to re-identify the pedestrian to be identified.

Abstract: The present invention provides a method and device for comparing similarities of high dimensional features of images, capable of improving the retrieval speed and retrieval precision in a similarity retrieval from massive images based on Locality Sensitive HASH (LSH) code. The method for comparing similarities of high dimensional features of images according to the present invention comprises: reducing dimensions of extracted eigenvectors of the images by the LSH algorithm to obtain low dimensional eigenvectors; averagely segmenting the low dimensional eigenvectors and establishing a segment index table; retrieving the segmented low dimensional eigenvector of a queried image from the segment index table to obtain a candidate sample set; and performing a similarity metric between a sample in the candidate sample set and the low dimensional eigenvector of the queried image.

Abstract: An article storage device includes a first conveying line, a second conveying line, and a temporary storage passage, wherein the first conveying line and the second conveying line ascend spirally around the temporary storage passage to form a double spiral structure, and two ends of the temporary storage passage respectively are in communication with the first conveying line and the second conveying line, the first conveying line conveys the article to the temporary storage passage, the temporary storage passage receives the article from the first conveying line and conveys the article to the second conveying line, and the second conveying line receives the article from the temporary storage passage and conveys the article.