Hybrid Orientation Based Human Limbs Motion Tracking Method

Abstract

One of the key technologies that lays behind the human–machine interaction and human motion diagnosis is the limbs motion tracking. To make the limbs tracking efficient, it must be able to estimate a precise and unambiguous position of each tracked human joint and resulting body part pose. In recent years, body pose estimation became very popular and broadly available for home users because of easy access to cheap tracking devices. Their robustness can be improved by different tracking modes data fusion. The paper defines the novel approach—orientation based data fusion—instead of dominating in literature position based approach, for two classes of tracking devices: depth sensors (i.e., Microsoft Kinect) and inertial measurement units (IMU). The detailed analysis of their working characteristics allowed to elaborate a new method that let fuse more precisely limbs orientation data from both devices and compensates their imprecisions. The paper presents the series of performed experiments that verified the method’s accuracy. This novel approach allowed to outperform the precision of position-based joints tracking, the methods dominating in the literature, of up to 18%.

Data Driven Models for Human Motion Prediction in Human-Robot Collaboration

“…Various effective fusion algorithms proposed in recent years have been applied to the field of human motion tracking and recognition. Glonek and Wojciechowski presented a data fusion method for human tracking, which can more accurately fuse the limb orientation data from two devices and compensate their imprecisions [43]. McIlwraith et al proposed a framework to describe posture evolution by fusing the environment and wearable sensing patterns [44].…”

Low Cost Inertial Sensors for the Motion Tracking and Orientation Estimation of Human Upper Limbs in Neurological Rehabilitation

“…Glonek and Wojciechowski [11] fuse the Kinect depth sensor and the IMU data to compensate for the limitations of both measurement devices. However, the use of the camera raises the concern on the privacy issues and may cause the patients unpleasant or uncomfortable during assessment.…”

Multi-camera Torso Pose Estimation using Graph Neural Networks

“…Among other relevant information such as gestures or facial expressions, people's position and orientation are among the most important cues that can help service and assistive social robots understand humans. A common application of human localisation and orientation is predicting intentions and movements in surveillance video feeds [1], [2]. An accurate localisation and orientation estimation are also crucial for human-aware navigation [3].…”

“…Although there is a considerable number of exceptions (e.g., [5]- [7]) orientation and other social cues are usually acquired using a two-stage pipeline: human body parts are detected as a first step and then passed as input to a second stage algorithm. This second algorithm is frequently implemented using basic trigonometry, considering the coordinates of the shoulders or the hips [1]. For instance, in [8], it is calculated using the cross-product of the vectors going from the head to the right and left sides of the hip, respectively.…”