JD has been granted a patent for a method and device to recognize videos using a pre-trained local and global diffusion model. The model learns spatial-temporal representations in the video based on diffusion between local and global representations. GlobalData’s report on JD gives a 360-degree view of the company including its patenting strategy. Buy the report here.

Smarter leaders trust GlobalData

Report-cover

Data Insights JD.com - Company Profile

Buy the Report

Data Insights

The gold standard of business intelligence.

Find out more

According to GlobalData’s company profile on JD, Automated guided vehicles was a key innovation area identified from patents. JD's grant share as of March 2024 was 32%. Grant share is based on the ratio of number of grants to total number of patents.

Video recognition method using pre-trained lgd model

Source: United States Patent and Trademark Office (USPTO). Credit: JD.com Inc

A recently granted patent (Publication Number: US11967134B2) discloses a method for video recognition using a pre-trained local and global diffusion (LGD) model. The method involves acquiring a video to be recognized and inputting it into the LGD model to determine the category of the video. The LGD model learns a spatio-temporal representation in the video by diffusing between local and global representations. The model consists of cascaded LGD modules, a local and global combination classifier, and a fully connected layer, with each LGD module comprising local and global paths that interact to describe local variation and holistic appearance at different spatio-temporal locations.

Furthermore, the patent details the process of learning local and global representations in parallel based on the video and cascaded LGD modules. The method involves segmenting the video into segments, selecting frames, and inputting them into LGD-2D modules to learn representations. Alternatively, the video can be segmented into segments and input into LGD-3D modules for parallel learning. The LGD-3D modules decompose three-dimensional learning into spatial convolutions and temporal operations. Additionally, the patent includes claims for a server and a computer-readable medium storing a program to implement the video recognition method using the LGD model, highlighting the potential applications and implementations of this innovative technology in various industries.

To know more about GlobalData’s detailed insights on JD, buy the report here.

Data Insights

From

The gold standard of business intelligence.

Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.

GlobalData

GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.

GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.