BaroPoser: Real-time Human Motion Tracking from IMUs and Barometers In Everyday Devices > 자유게시판

• 목록
• 답변
• 글쓰기
• 게시판 리스트 옵션
  • 수정
  • 삭제

In recent years, monitoring human motion using IMUs from everyday devices similar to smartphones and smartwatches has gained rising recognition. However, as a result of sparsity of sensor measurements and the lack of datasets capturing human movement over uneven terrain, current strategies usually battle with pose estimation accuracy and are usually limited to recovering movements on flat terrain solely. To this finish, we present BaroPoser, iTagPro bluetooth tracker the primary technique that combines IMU and barometric information recorded by a smartphone and a smartwatch to estimate human pose and global translation in real time. By leveraging barometric readings, we estimate sensor height adjustments, which give precious cues for both enhancing the accuracy of human pose estimation and predicting international translation on non-flat terrain. Furthermore, we suggest a local thigh coordinate frame to disentangle native and international motion input for higher pose representation studying. We evaluate our methodology on each public benchmark datasets and actual-world recordings. Quantitative and qualitative outcomes display that our strategy outperforms the state-of-the-art (SOTA) strategies that use IMUs only with the same hardware configuration.

Human motion capture (MoCap) is a protracted-standing and challenging problem in laptop graphics and imaginative and ItagPro prescient. It goals to reconstruct 3D human physique movements, with many purposes in film manufacturing, gaming, and AR/VR. Although imaginative and prescient-primarily based strategies (Peng et al., 2021; Lin et al., 2024; Wang et al., iTagPro shop 2024; Xiu et al., 2024) have made vital progress on this field, they always require digicam visibility and are sensitive to occlusions and lighting conditions. Some works have targeted on monitoring human movement through accelerations and iTagPro product rotations recorded by physique-worn Inertial Measurement Units (IMUs), which overcome the aforementioned limitations. Commercial solutions on this category require 17 or more IMUs, ItagPro which will be intrusive and time-consuming for utilization. Recent studies (Huang et al., 2018; Yi et al., 2021, 2022; Jiang et al., 2022; Van Wouwe et al., 2023; Yi et al., 2024; Armani et al., 2024) have diminished the number of IMUs to six or fewer, striking a balance between accuracy and practicality.

However, these strategies still require specialized IMU sensors, limiting their application in on a regular basis life. To deal with this problem, ItagPro some studies (Mollyn et al., 2023; Xu et al., 2024) leverage the IMUs already accessible in on a regular basis devices (comparable to phones, watches, wristbands, and earbuds) for human movement capture. These strategies outline a set of typical machine placement locations like the pinnacle, wrists, or pockets, and use up to three of them to estimate human pose and global translation by way of neural networks. However, the accuracy of their motion estimation remains restricted, as the issue is inherently below-constrained attributable to sparse and noisy IMU measurements obtainable in everyday settings. This makes it troublesome to precisely get better either local physique poses or international translations. In this paper, we current BaroPoser, the primary approach that fuses IMU and barometric data from one smartwatch (worn on one wrist) and one smartphone (positioned in the thigh pocket of the alternative aspect) to estimate full-physique motions in actual time.

Along with IMU data, which has been broadly used for MoCap, we suggest to incorporate barometric readings from built-in sensors in on a regular basis devices equivalent to smartphones and smartwatches. These readings provide information about absolute altitude, offering an extra function to improve the accuracy of each pose and global translation estimation. Such vertical awareness is particularly invaluable in applications equivalent to AR/VR and health monitoring, where altitude-sensitive actions like stair climbing, squats, and jumps are common. Moreover, to higher exploit human motion priors on this setting with only two sensors, we introduce a thigh coordinate system to symbolize local body poses, which helps to decouple the local and global motions. Specifically, we outline an area coordinate frame for the sensor on the thigh and treat it as the basis coordinate frame of the human local pose. Then, iTagPro both the input and output of the pose estimation network are represented in this root coordinate frame. On this method, the worldwide and native movement info recorded by the sensors is disentangled naturally because the thigh sensor data the worldwide motion information whereas the wrist sensor records the local one.