Kia had 71 patents in artificial intelligence during Q1 2024. Kia Corp filed patents in Q1 2024 for various innovative technologies including an apparatus for selecting training images for deep learning models, an object recognition apparatus using 2D and 3D images, a free motion headform impact prediction device using artificial intelligence, a method for training machine reading comprehension models, and an apparatus for diagnosing vehicle component faults using machine learning models. These patents showcase Kia‘s focus on advanced technologies for image processing, artificial intelligence, and fault diagnosis in vehicles. GlobalData’s report on Kia gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Kia grant share with artificial intelligence as a theme is 38% in Q1 2024. Grant share is based on the ratio of number of grants to total number of patents.
Recent Patents
Application: Apparatus for selecting a training image of a deep learning model and a method thereof (Patent ID: US20240104901A1)
The patent filed by Kia Corp. discloses an apparatus and method for selecting a training image for a deep learning model. The apparatus includes an input device that receives a simulation image and object information from a simulation tool, as well as a training image from an image conversion device. A controller then detects the similarity between the object structures in the simulation and training images to determine the validity of the training image based on this similarity. The controller can determine the validity of the training image by comparing the detected similarity to a threshold value, storing valid images, and considering similarities in multiple objects.
Furthermore, the method involves receiving a simulation image and object information, receiving a corresponding training image, detecting the similarity between object structures, and determining the validity of the training image based on this similarity. The method includes steps such as storing valid images, considering similarities in multiple objects, and utilizing a structural similarity index measure (SSIM) to assess the similarity between images. Additionally, the method involves generating simulation images based on various scenarios and converting them into training images using a generative adversarial network (GAN) through the image conversion device.
To know more about GlobalData’s detailed insights on Kia, 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.