Hinge Health‘s patent involves a system for estimating 3D joint locations from 2D images using convolutional neural networks. By combining 2D joint heatmaps with image segments, the technology accurately determines the depth of skeleton joints, providing a comprehensive 3D joint position estimation. GlobalData’s report on Hinge Health gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on Hinge Health, AI-assisted medical imaging was a key innovation area identified from patents. Hinge Health's grant share as of January 2024 was 12%. Grant share is based on the ratio of number of grants to total number of patents.
Estimating 3d joint location from 2d image of skeleton
A recently granted patent (Publication Number: US11875529B2) discloses a non-transitory medium with instructions that, when executed by a computing system's processor, enable the system to estimate three-dimensional (3D) positions of joints in a person's image. The method involves obtaining a two-dimensional (2D) image, determining a 2D joint skeleton and heatmap, applying them to a convolutional neural network with 3D convolutional layers to expand depth resolution, and combining the estimated depth with the 2D joint skeleton to construct a 3D joint skeleton. Additional operations include identifying the person's segment in the image and utilizing 2D and 3D convolutional layers for processing.
Furthermore, the patent describes a method for estimating 3D joint positions by analyzing an image segment containing the joints. This method involves creating a 2D joint heatmap based on a 2D joint skeleton, applying it to a convolutional neural network with 3D convolutional layers to expand depth resolution, and combining the 2D location with the estimated depth to generate the 3D position for each joint. The method also includes utilizing max pooling layers to reduce 2D resolution and reshape layers to adjust the depth dimension. The patent also covers a computing system that implements these instructions to estimate 3D joint positions in an object, particularly a human, through image analysis and neural network processing, emphasizing the combination of 2D location and estimated depth for accurate 3D positioning.
To know more about GlobalData’s detailed insights on Hinge Health, 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.