Noitom
More actions
| Noitom | |
|---|---|
| Information | |
| Industry | Motion capture |
| Headquarters | Beijing, China |
| Products | Perception Neuron, Hi5 VR gloves, Project Alice |
| Website | noitom.com |
Noitom is a Chinese developer of motion capture hardware and software. Its products include the Perception Neuron family of wearable inertial systems, Hi5 gloves for hand and finger input, and Project Alice, a commercial virtual reality system that combines optical and inertial tracking. The company identifies its Beijing business as Beijing Noitom Technology Ltd. (北京诺亦腾科技有限公司).[1][2]
Noitom equipment has two documented uses in XR: recording performances for animated characters, and supplying live body tracking or hand movement to an interactive application. Inertial motion capture can supply joint and finger motion, while a separate positional reference can locate the performer or hand in the same room coordinates as a headset.[3][4]
Company and product history
Noitom's current Chinese website describes Beijing Noitom Technology Ltd. as established in 2012 and headquartered in Beijing. A company-issued Hi5 announcement from January 2017 instead gave 2011 as the founding year of Noitom Ltd.; the two company accounts use different dates.[1][5]
| Date | Documented development |
|---|---|
| 2014 | Perception Neuron's Kickstarter campaign began on 8 August and ended in September. Contemporary reporting recorded nearly US$572,000 raised against a US$250,000 goal. The proposed sensor packages supported body movement and individual fingers; the campaign's announced delivery date was a forecast, not evidence of shipment.[6] |
| January 2016 | Noitom announced Project Alice in China as a commercial VR system. UploadVR's corrected report described its initial focus as business customers in China.[2] |
| January 2017 | Noitom presented the Hi5 VR Glove at CES with HTC Vive. Its announcement identified Tristan Dai, then Noitom's CTO, as Hi5's creator; an independent hands-on report documented a Vive Tracker mounted at the wrist.[5][7] |
| February 2018 | Perception Neuron 2.0 was reported as a hardware refresh of the original system, with revised straps, cables, connectors and sensor housings. The accompanying accessory ecosystem accommodated both 1.0 and 2.0 sensors.[8] |
| 2025 product revisions | Noitom's current PN 3 and PN Studio pages describe upgrades in 2025 to versions marketed as resistant to magnetic interference. These are later product revisions than the equipment described in the early Kickstarter coverage.[9][10] |
Motion capture systems
Perception Neuron uses inertial measurement units worn on body segments. Noitom's current PN 3 description lists a gyroscope, accelerometer and magnetometer in each nine-axis sensor, with an attitude and heading reference system that produces orientation data. Axis Studio records or broadcasts the resulting motion and exports BVH or FBX files.[9]
The current catalog contains several distinct systems. The descriptions below reflect the manufacturer's published configurations, rather than independently measured performance.[9][10][11][12]
| System | Published configuration and scope |
|---|---|
| PN 3, 2025 anti-magnetic revision | Wireless inertial capture of body and fingers. The current page describes 27 sensor nodes distributed over the body and hands, a receiver and charging equipment.[9] |
| PN Studio, 2025 anti-magnetic revision | Wireless inertial capture with body sensors and separate gloves. The published standard package lists 17 body sensors, one spare and a pair of PN S gloves. Noitom describes an optional customized upgrade to optical-inertial tracking.[10] |
| PN Link | A wired inertial system with Power over Ethernet or a direct USB Type-C power connection. Noitom describes the addition of optical trackers for a hybrid mode that supplies absolute position for tracked end points and props.[11] |
| PN Hybrid | An optical-inertial platform for rigid-body spatial tracking. Its documentation distinguishes optical position measurements from inertial orientation measurements and describes six degrees of freedom output.[12] |
The word "wireless" also needs a generation-specific reading. In the early Perception Neuron kit reviewed in 2016, small sensors connected through cables on the body, while the system could communicate with the computer by USB or Wi-Fi. That arrangement differs from later products with individually wireless sensor modules.[3][9]
For avatar animation, captured motion must match the character's skeleton or undergo retargeting. UploadVR's 2016 developer review described a workflow of capture, FBX export, engine import and retargeting, and demonstrated live input through a Unity plugin. The author reported setup and calibration work, drift and animation cleanup needs in that early kit; those observations concern the tested hardware and workflow.[3]
Hi5 hand input
The original Hi5 VR Glove combined inertial sensing of the hand and fingers with an external HTC Vive Tracker or supported Vive controller for absolute hand and forearm position. Noitom's version 1.4 guidelines require the external tracking device for positioning and calibration, describe vibration feedback, and provide Unity and Unreal Engine SDKs. The guidelines also warn about magnetic or ferrous objects and require recalibration when the optical tracking system moves or the wearer changes.[4]
At CES 2017, Ben Lang's hands-on report described pointing, manipulating virtual objects and individual finger motion. He also reported that horizontal finger motion did not appear to be tracked in that demonstration. This was an observation of the CES prototype, not a measurement of every later Hi5 version.[7]
Hi5 2.0 has a separate software and plugin page. The published PC offering includes basic hardware-management and calibration SDKs plus interaction SDKs for Unity and Unreal on SteamVR. The page lists Unreal versions 4.26, 4.27, 5.0, 5.1 and 5.2 and states that it has no plan for later-version support. It separately lists Unity plugins for Pico Neo 3 and Pico 4 on Android. These documented combinations do not establish compatibility with every headset or engine release.[13]
XR production and interaction
Performance capture
In January 2016, Cloudhead Games described using Perception Neuron with HTC Vive for performances in The Gallery. The actor could see the virtual scene while the suit recorded movement; custom Unity rigs let the scene respond to the performance and supplied virtual cues. This was a content-production use of motion capture rather than a requirement that players wear a suit.[14]
Project Alice
Project Alice is Noitom's commercial system for multiple participants and tracked physical objects in a shared VR space. Noitom's documentation describes optical-inertial tracking, computing and rendering components, and software and hardware management. Its stated application types include simulation laboratories, training and theme-park experiences.[15]
UploadVR's April 2016 SVVR report documented a demonstration with five Oculus Rift DK2 headsets, separate rendering PCs and a central server. The demonstration used OptiTrack cameras and tracked props, including a basketball, so participants could pass an object between physical and virtual space. Its five-person configuration and announced US expansion plans describe that demonstration and announcement; they do not define all subsequent Project Alice installations.[16]
Research systems
The 2016 GraphiteVR preprint by Sam Royston, Connor DeFanti and Ken Perlin described collaborative social-network visualization using Samsung Gear VR, Holojam, OptiTrack and Perception Neuron. The researchers used a single-arm model to obtain hand and finger motion, and added optical markers to position the glove because the inertial system lacked an absolute positional reference. Gestures moved and scaled the graph. The paper documents a prototype architecture rather than a comparative user study or a commercial headset specification.[17]
The 2023 Robotics: Science and Systems paper on RFUniverse documented another use: an interface to a multiphysics robotics simulator through SteamVR, demonstrated with an HTC Vive and a Noitom Hi5 glove. The paper's simulator experiments evaluate simulation tasks; their results are not measurements of the glove's tracking accuracy.[18]
Published validation and limits
Researchers have evaluated Perception Neuron by comparing joint-angle data with camera-based motion capture. These studies test a particular sensor configuration, calibration procedure, software pipeline and set of movements. They measure agreement of body-motion estimates, which is different from measuring headset positional tracking or end-to-end VR latency.[19][20]
| Study | Tested configuration and tasks | Findings and limits |
|---|---|---|
| Choo, Chow and Komar (2022) | Ten participants; 17 IMUs and Axis Neuron over USB, compared with an eight-camera Vicon system at 120 Hz. Tasks included walking, jogging and stationary or moving floorball wrist shots. | Most raw joint-angle waveform comparisons had root mean square error below 4°, with larger errors for shoulder abduction/adduction and ankle flexion/extension during jogging. The authors distinguished normalized movement patterns from raw joint angles and cautioned against treating the latter as interchangeable. Recordings lasted 20-30 seconds before recalibration, limiting conclusions about prolonged drift.[19] |
| Shuai and colleagues (2022) | Twenty healthy adults tested on two days; seven wireless IMUs with Axis Studio, compared with eight OptiTrack cameras at 100 Hz. Tasks were gait, squat, single-leg squat, side and forward lunges, and countermovement jump. | Joint-angle waveform RMSE before offset correction ranged from 3.57° to 13.14°. Agreement varied with joint and plane of motion, and the authors discussed differences between biomechanical models. The sample did not establish validity for patients or elite athletes.[20] |
These results concern the equipment and processing described in the papers. Neither study evaluates the 2025 anti-magnetic PN 3 or PN Studio revisions, PN Link, or PN Hybrid. Newer manufacturer descriptions therefore require separate evaluation before numerical results from an older system can be applied to them.[19][20][9][10][11][12]
References
- ↑ 1.0 1.1 "关于我们". Noitom. https://www.noitom.com.cn/about.html. Retrieved 2026-10-12.
- ↑ 2.0 2.1 Will Mason (2016-01-15). "Perception Neuron Creator's 'Project Alice' Brings More Than Just Your Body into the Game". UploadVR. https://www.uploadvr.com/noitom-project-alice-announced-objects/. Retrieved 2026-10-12.
- ↑ 3.0 3.1 3.2 Sky Nite (2016-04-18). "Perception Neuron Review: In-Depth With The $1,500 Motion Capture Suit". UploadVR. https://www.uploadvr.com/perception-neuron-review/. Retrieved 2026-10-12.
- ↑ 4.0 4.1 "Noitom Hi5 VR Glove User Guidelines, Ver. 1.4". Noitom. 2018-04-13. https://shopcdn.noitom.com.cn/hi5/HI5_user_guidelines_V1.4.pdf. Retrieved 2026-10-12.
- ↑ 5.0 5.1 Noitom Ltd. PR (2017-01-05). "Noitom Announces VR Motion Capture Glove During CES 2017". PRWeb. https://www.prweb.com/releases/noitom_announces_vr_motion_capture_glove_during_ces_2017/prweb13954839.htm. Retrieved 2026-10-12.
- ↑ Ben Lang (2014-09-13). "Perception Neuron More Than Doubles Kickstarter Goal for VR Input Suit at $572,000". Road to VR. https://roadtovr.com/perception-neuron-vr-input-suit-kickstarter-double-goal-ending-572000/. Retrieved 2026-10-12.
- ↑ 7.0 7.1 Ben Lang (2017-01-06). "Hands-on: Noitom’s Hi5 VR Glove Brings Compelling Finger Tracking to the Vive". Road to VR. https://roadtovr.com/noitom-hi5-vr-glove-htc-vive-finger-tracking-hands-on/. Retrieved 2026-10-12.
- ↑ Scott Hayden (2018-02-05). "Noitom Releases Perception Neuron 2.0 Motion Capture System". Road to VR. https://roadtovr.com/noitom-releases-perception-neuron-2-0-motion-capture-system/. Retrieved 2026-10-12.
- ↑ 9.0 9.1 9.2 9.3 9.4 9.5 "PERCEPTION NEURON 3". Noitom. https://www.noitom.com.cn/perception-neuron-3-pro.html. Retrieved 2026-10-12.
- ↑ 10.0 10.1 10.2 10.3 "PERCEPTION NEURON STUDIO". Noitom. https://www.noitom.com.cn/perception-neuron-studio.html. Retrieved 2026-10-12.
- ↑ 11.0 11.1 11.2 "PERCEPTION NEURON LINK". Noitom. https://www.noitom.com.cn/perception-neuron-link.html. Retrieved 2026-10-12.
- ↑ 12.0 12.1 12.2 "PERCEPTION NEURON HYBRID". Noitom. https://www.noitom.com.cn/perception-neuron-hybrid.html. Retrieved 2026-10-12.
- ↑ "Hi5 2.0 VR 交互手套软件和插件下载". Noitom. https://shopcdn.noitom.com.cn/article/43.html. Retrieved 2026-10-12.
- ↑ Joe Durbin (2016-01-14). "Cloudhead Games and Perception Neuron Redefine Motion Capture For VR". UploadVR. https://www.uploadvr.com/cloudhead-motion-capture-gallery/. Retrieved 2026-10-12.
- ↑ "Project Alice". Noitom. https://www.noitom.com.cn/project-alice.html. Retrieved 2026-10-12.
- ↑ Joe Durbin (2016-04-29). "This Is What A $100,000 VR Setup Looks Like". UploadVR. https://www.uploadvr.com/100000-vr-system-looks-like/. Retrieved 2026-10-12.
- ↑ Sam Royston, Connor DeFanti and Ken Perlin (2016-04-27). "GraphiteVR: A Collaborative Untethered Virtual Reality Environment for Interactive Social Network Visualization". arXiv, 1604.08239. https://arxiv.org/pdf/1604.08239. Retrieved 2026-10-12.
- ↑ Haoyuan Fu, Wenqiang Xu, Ruolin Ye, Han Xue, Zhenjun Yu, Tutian Tang, Yutong Li, Wenxin Du, Jieyi Zhang and Cewu Lu (2023). "Demonstrating RFUniverse: A Multiphysics Simulation Platform for Embodied AI". Robotics: Science and Systems 2023. https://www.roboticsproceedings.org/rss19/p087.pdf. Retrieved 2026-10-12.
- ↑ 19.0 19.1 19.2 Corliss Zhi Yi Choo, Jia Yi Chow and John Komar (2022-01-21). "Validation of the Perception Neuron system for full-body motion capture". PLOS ONE, 17(1), e0262730. doi:10.1371/journal.pone.0262730. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0262730. Retrieved 2026-10-12.
- ↑ 20.0 20.1 20.2 Zhenyu Shuai, Anqi Dong, Haoyang Liu and Yixiong Cui (2022-01-23). "Reliability and Validity of an Inertial Measurement System to Quantify Lower Extremity Joint Angle in Functional Movements". Sensors, 22(3), 863. doi:10.3390/s22030863. https://pmc.ncbi.nlm.nih.gov/articles/PMC8838175/. Retrieved 2026-10-12.