Naver had 33 patents in artificial intelligence during Q2 2024. Naver Corp filed patents related to training systems using pairs of images to adjust model parameters, a method for unsupervised pre-training and fine-tuning of models for geometric vision tasks, an information retrieval training system without labels, and a language learning service using target images and language learning information. GlobalData’s report on Naver gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Naver had no grants in artificial intelligence as a theme in Q2 2024.
Recent Patents
Application: Unsupervised pre-training of geometric vision models (Patent ID: US20240144019A1)
The patent filed by Naver Corp. describes a training system that involves constructing pairs of images of different portions of humans captured at different times and viewpoints, inputting these pairs into a model, generating reconstructed images based on the input pairs, and adjusting model parameters based on the differences between the reconstructed images and predetermined images. The method involves unsupervised pre-training of a machine learning model, initializing task-specific encoder parameters based on the pre-trained model, and fine-tuning the model for downstream geometric vision tasks. The system also includes a masking module to mask pixels of human portions in images, allowing for selective adjustment of model parameters based on differences between reconstructed and predetermined images.
The training system further allows for the construction of pairs of images of different human portions captured at different times and viewpoints, with adjustments made to model parameters based on differences between reconstructed and predetermined images. The system accounts for variations in ethnicity, age, gender, pose, background, clothing texture, and body shape between different humans in the images. Additionally, the training method involves fine-tuning the model for specific tasks such as determining mesh of hand or body surfaces, coordinates of body surfaces, three-dimensional pose estimation, or mesh reconstruction from image pairs. The system aims to improve the accuracy and efficiency of machine learning models for geometric vision tasks through pre-training, parameter adjustment, and fine-tuning processes.
To know more about GlobalData’s detailed insights on Naver, 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, 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.
