Markerless motion capture for 3D human model animation using depth camera

Abstract

3D animation is created using keyframe based system in 3D animation software such as Blender and Maya. Due to the long time interval and the need of high expertise in 3D animation, motion capture devices were used as an alternative and Microsoft Kinect v2 sensor is one of them. This research analyses the capabilities of the Kinect sensor in producing 3D human model animations using motion capture and keyframe based animation system in reference to a live motion performance. The quality, time interval and cost of both animation results were compared. The experimental result shows that motion capture system with Kinect sensor consumed less time (only 2.6%) and cost (30%) in the long run (10 minutes of animation) compare to keyframe-based system, but it produced lower quality animation. This was due to the lack of body detection accuracy when there is obstruction. Moreover, the sensor's constant assumption that the performer's body faces forward made it unreliable to be used for a wide variety of movements. Furthermore, standard test defined in this research covers most body parts' movements to evaluate other motion capture system.

Classification of lower limb rehabilitation exercises with multiple and individual inertial measurement units

“…where normal vector of a hyperplane is represented by ′𝜔′ and an offset from the origin along this normal vector is represented by 'd'. Equation (3) can be expressed using the Karush-Kuhn-Tucker (KKT) conditions having Lagrangian multipliers, as (5):…”

“…For example, a time-consuming data processing with expertise to interpret the processed dataset and comment on the output is required for 3D motion capture systems. Although a fast and economical solution for motion capture device could be a single Microsoft Kinect v2 sensor, it has many limitations that made it unreliable to be used for wide variety of movements [5]. Visual assessment contains, generally, force plates, electromyography (EMG), footswitches, and motion capture systems.…”

A robust method for VR-based hand gesture recognition using density-based CNN

“…There are two kinds of motion sensors, wearable sensors and camera-based sensors. In case using a camera-based sensor, from frames captured over time, the displacement of human joints position will be considered as human motion [9][10][11][20][21][22][23][24]. Motion with certain patterns will be understood as a gesture.…”

“…It needs depth information to extract a feature vector from the palm area. Yang used HMM [26], Molchanov used HOG [11], some others used spatio-temporal feature [20,23,[27][28][29] and others used motion feature [9,24,30]. Yet, to recognize more various hand motion gestures, skeleton-based data is better [3,22,24,28].…”