RTS-Tenoomi, as the next-generation computing platform for virtual reality (VR), has seen not only major companies providing platforms and resources joining the VR market but also high-quality enterprises deeply involved in industry applications gradually emerging. Interaction within the same scene for multiple people has become a standard feature of VR solutions, and the increasing number of successful VR cases in various vertical industries is a strong testament to the positive development trend of the VR industry.
However, most domestic multi-person interactive large-space VR solutions currently can only track the information of players’ hands and heads. In the shared virtual space, players can only see each other’s accurate head and hand positions but lack complete body posture synchronization. This significantly reduces the immersion and interaction authenticity of VR in some industry applications, limiting user experience and industry development. It is reported that to adapt to the development trend of VR industry applications and to solve the limitations of capture, Shenzhen Realis Multimedia Technology Co., Ltd., a leading domestic motion capture company, has officially announced the upcoming launch of the software Tenoomi and Rovonga, two new members for multi-person full-body motion capture: RTS-Tenoomi is a full-body motion capture software based on rigid bodies, suitable for real-time human-computer interaction in VR virtual reality, virtual anchors, virtual shooting, and other occasions. It was officially released and commercially available in May 2018. The software’s biggest feature is that, in addition to basic full-body motion capture functions, its SDK will seamlessly integrate with major VR engines (Unreal Engine 4, Unity, etc.). The performer’s body movements will be directly reflected within the VR engine, interacting with the models and interaction logic within the scene with zero latency. When Tenoomi is working, it requires rigid bodies to be placed on the performer’s head, hands, feet, and back, receiving rigid body data (position and orientation information) from the performer’s body transmitted by Realis’s tracking and positioning software RTS-Tracker, obtaining the performer’s limb positions, and then combining IK technology and HSM models for full-body skeletal resolution to estimate the performer’s full-body motion skeletal posture.It is reported that Tenoomi has the following features: High-efficiency natural data calculation RTS-Tenoomi can obtain high-precision rigid body data from RTS-Tracker with just one click. Each person only needs six rigid body data to perform full-body skeletal calculation, which results in less data processing and thus enables low-latency, precise real-time interaction for multiple people. Traditional motion capture often requires performers to wear tight-fitting clothing with dozens of reflective points or inertial wearable devices with nodes and cables all over their body.
The purpose is to achieve high fluidity in motion settlement. However, considering the efficiency and wearability that VR applications must satisfy, VR users cannot go through the cumbersome process of dressing or changing tight clothes before each experience. Therefore, Tenoomi has been designed to perform skeletal calculation and redirection without requirements for user body type or redirected character models, meaning that with just six rigid body data, regardless of the user’s body size or height, natural and smooth full-body skeletal postures can be calculated and redirected to virtual character models of different proportions and skeletons, demonstrating strong practicality.Support for online multiplayer real-time interaction is a feature that is rarely seen in current full-body motion capture software abroad. Although they can achieve smooth full-body motion capture, most focus on online recording for film, sports research, and offline action editing, with fewer applications for real-time interaction in online multiplayer full-body motion capture. The reason is that the data processing volume in existing full-body motion capture processes is huge, and capturing multiple people simultaneously in real-time can easily cause severe delays in VR, which is unacceptable to users. In the process of online multiplayer full-body interaction, there are the following issues: it is easy to achieve full-body motion capture for a single person, but it is very difficult for multiple people in the same scene because the number of rigid bodies for full-body motion capture increases linearly with each additional person. How to process data, how to avoid and estimate occlusions, and how to perform skeletal calculation to ensure real-time and high precision in interaction are all challenges. To address these issues, RTS-Tenoomi launched by Rivalight has made a lot of optimizations in data processing. Firstly, the number of rigid bodies worn by users has been reduced to six, which reduces the source data volume. However, to ensure the effect of full-body skeletal estimation, Tenoomi applies a proprietary skeletal posture algorithm from Rivalight for skeletal calculation. This algorithm can successfully optimize occlusion issues and perform precise, smooth, and natural full-body skeletal estimation even with a small amount of source data, thus ensuring real-time full-body interaction for multiple people online simultaneously, which is very rare in the industry.
Tenoomi boasts a user-friendly UI interface, providing a pleasant interaction experience. Users can easily engage with the system without the need for special attire, simply by wearing basic gloves or foot covers. This convenience facilitates the experience. Moreover, Tenoomi can accurately perform full-body calculations without changing the solver, even when user sizes vary slightly, simplifying the user operation process and enhancing the overall experience.
Unreal Engine and Unity engines are seamlessly integrated with Tenoomi, which also includes its dedicated UE4/Unity SDK for release. Regardless of the engine or version developers use, Tenoomi offers seamless support and integration. Developers no longer need to edit their own IK blueprints or C++ code for full-body motion capture in VR; Tenoomi’s SDK handles all posture and motion expressions in VR. The SDK is also provided free of charge to development teams. With the software activated, multiple virtual characters can engage in natural and realistic full-body motion interactions in VR.
Based on motion capture technology, Tenoomi’s skeletal assessment can be widely applied across various industries in VR scenarios, such as VR large space game venues, military training, professional training, and rehabilitation training. Through posture synchronization in the same scene with multiple people, more realistic motion training, escape routes, simulation training, and posture analysis can be achieved in VR, becoming a deeper and more natural means of human-computer interaction in virtual environments.
Let’s also discuss Rovonga, another full-body motion capture ‘new member’ that Realis is preparing to launch. Rovonga is a full-body motion capture software based on a dot-based system, capable of capturing high-precision, clean data with low loss rates for multiple people’s full-body motion data. It is mainly used in the film industry for high-fluidity animation production, scientific research, human body analysis, and sports industries. It is expected to be launched for the first time at the Siggraph Canada exhibition in August. According to Realis insiders, these two full-body motion capture software will complement each other, playing different roles in various industry application scenarios.
The goal of RTS-Rovonga full-body motion capture is to accompany the rise of entertainment and consumer market demands, with the industry application demands of virtual reality also releasing tremendous potential. An increasing number of vertical industry demands and increasingly professional projects require teams with technical backgrounds to support.
Riitek CEO Xu Qiuzi stated: “Riitek has been committed to optical motion capture and human-computer interaction multimedia technology. The full-body motion capture system we are now officially launching is dedicated to providing more users with a cost-effective, fully domestically developed, and professionally supported motion capture solution. In the virtual world, the quality of full-body motion capture for virtual avatars signifies that we have achieved hyper-realistic virtual humans who will move and interact more authentically in their world. Previously, achieving the same level of realism required a significant amount of time and higher costs. Riitek’s full-body motion capture solution is crucial for next-generation projects in film and television, military, VR development, sports, and industrial simulation. As VR technology evolves, we are moving towards the perfect world we once dreamed of. Future VR experiences are bound to head in the direction of increasingly realistic and naturally integrated human-computer interaction worlds. Motion capture technology is a key interactive technology in future VR technology and has been listed by the National Ministry of Industry and Information Technology as one of the five key technologies in VR. In the future, we will not be satisfied with just sitting in front of computers and televisions operating with keyboards and mice; we aspire to have a full-dimensional immersive interaction method that allows us to travel through different worlds and across different times and spaces.
