Patents Assigned to VISENZE PTE LTD
-
Patent number: 11869228Abstract: A method of generating an image recognition model for recognising an input image and a system thereof are provided. The method includes appending at least one feature extraction layer to the image recognition model, extracting a plurality of feature vectors from a set of predetermined images, grouping the plurality of feature vectors into a plurality of categories, clustering the plurality of feature vectors of each of the plurality of categories into at least one cluster, determining at least one centroid for each of the at least one cluster, such that each of the at least one cluster comprises at least one centroid, such that each of the at least one centroid is represented by a feature vector, generating a classification layer based on the feature vector of the at least one centroid of the plurality of categories, and appending the classification layer to the image recognition model. In addition, a method of classifying an input image and a system thereof are provided.Type: GrantFiled: November 6, 2020Date of Patent: January 9, 2024Assignee: VISENZE PTE LTDInventors: Guangda Li, Thanh Tung Todd Cao, Zhenhua Wang, Xin Ji
-
Patent number: 10949460Abstract: According to various embodiments, the present invention provides a product indexing method including receiving product data of a product, the product data comprising at least an image, such that the image includes a product image of the product and a context image which provides the context that the product may be in, such that the context image includes at least one non-product image. Product indexing method further includes identifying the product image, identifying the context image from the image, verifying the product in the product image based on the context image, extracting the product image and indexing the product image. According to various embodiment, the present invention provides a product indexing system for the product indexing system.Type: GrantFiled: February 24, 2015Date of Patent: March 16, 2021Assignee: VISENZE PTE LTDInventors: Guangda Li, Zheng Song, Fan Xu
-
Patent number: 10779037Abstract: Present example embodiments relate generally to systems, devices, and methods for managing video content. The method may include receiving an input video having one or more video frames. The method may also include selecting a video frame from the input video that displays a target product and/or other feature. The method may also include generating an alternative representation of the target product and/or other feature displayed in the selected video frame. The method may also include searching a content database for media content that is likely to pertain to the target product and/or other feature displayed in the selected video frame. The searching of the content database may be performed using the alternative representation of the target product and/or other feature. The method may also include associating a most likely matching media content found, in the searching, to the selected video frame.Type: GrantFiled: January 30, 2019Date of Patent: September 15, 2020Assignee: VISENZE PTE LTDInventors: Guangda Li, Zheng Song, Fan Xu
-
Patent number: 10747826Abstract: A clothing search system provides a clothing search to users using a component-based image search. Retailer catalogs are analyzed to determine clothing components within clothing images. Features associated with the components are determined. When a user requests a clothing search, the clothing search system selects clothing based on the components and features requested by the user. The user may also provide an image to the clothing search system. The clothing search system determines components and features of the image and identifies clothing with matching components.Type: GrantFiled: October 11, 2017Date of Patent: August 18, 2020Assignee: VISENZE PTE. LTDInventors: Shuicheng Yan, Zheng Song, Si Liu
-
Patent number: 10715862Abstract: Media content relevant to an input video is identified by analyzing video frames in the input video to detect if any of the video frames contain a target product. One or more video frames found to be containing a detected target product is selected and a product thumbnail is generated for each of selected the video frames. At least one product thumbnails is selected and a video product visual index is generated for each of the selected product thumbnail. Relevant media content is then identified for each of the product thumbnails by comparing the video product visual index of each of the selected product thumbnail with a plurality of media content visual index in a media content database. Each media content visual index is associated with one media content in the media content product database.Type: GrantFiled: February 24, 2015Date of Patent: July 14, 2020Assignee: VISENZE PTE LTDInventors: Guangda Li, Zheng Song, Fan Xu
-
Publication number: 20180032545Abstract: According to various embodiments, the present invention provides a product indexing method including receiving product data of a product, the product data comprising at least an image, such that the image includes a product image of the product and a context image which provides the context that the product may be in, such that the context image includes at least one non-product image. Product indexing method further includes identifying the product image, identifying the context image from the image, verifying the product in the product image based on the context image, extracting the product image and indexing the product image. According to various embodiment, the present invention provides a product indexing system for the product indexing system.Type: ApplicationFiled: February 24, 2015Publication date: February 1, 2018Applicant: VISENZE PTE LTDInventors: Guangda Li, Zheng Song, Fan Xu