- Acquire | MotionDecode
RoboDecode
MotionDecode is a high‑quality motion data system jointly built by CHINGMU and EnDecodeX for Embodied AI training, robot learning, and simulation validation. It covers a full range of data modalities – body, hand, object, video, environment, and task labels – and delivers end‑to‑end data services spanning real‑world motion acquisition, multimodal synchronization, quality screening, robot retargeting, and customized data delivery.
Table tennis
Dance
Basketball
Lifting and moving
Parkour
1,000 Hours of Data – Free to AccessThe "MotionDecode Data Open Program" is now officially launched. The first 1,000 hours of high-quality human motion data are open for free application via QR code – enabling teams to quickly kickstart Embodied AI training projects and accelerate algorithm iteration and validation.Real‑World Motion Data FoundationPowered by high‑precision optical mocap data, delivering ±0.01mm sub‑mm accuracy with drift‑free performance. With 3,000 hours of high‑quality human motion data already available, it provides a solid and reliable data foundation for Embodied AI training.
Multimodal Synchronized AcquisitionCovering multimodal data – body, hand skeletal motion, tactile pressure, first-person view (Ego), third-person view (Exo), object 6D pose, and environmental scenes – with precise temporal synchronization across all sources to build a complete motion data ecosystem.Fine Dexterous Hand Motion DataFeaturing joint angle error as low as 0.1°, with synchronized multimodal acquisition of tactile and pose data. Full hand coverage – five fingers plus wrist – enabled by optical‑inertial fusion technology for robust anti‑occlusion performance, accurately capturing a wide range of fine hand manipulations.
Multi-Format Flexible CompatibilitySupporting 10+ standard data formats including BVH, FBX, and CSV, with native compatibility for mainstream robot simulation environments such as MuJoCo, Isaac, and URDF. Easily integrates with diverse development platforms and toolchains for seamless incorporation into existing workflows.Standardized Data Production WorkflowA 7‑step standardized data production workflow covering task planning, hardware deployment, multimodal data acquisition, real‑time processing, QC & repair, packaging & upload, through to task completion. End‑to‑end quality control ensures stability and consistency in data delivery.
Custom Data ProductionSupporting full‑workflow customization across scenario design, robot platform adaptation, motion categories, and modality combinations – from requirement breakdown to asset delivery. End‑to‑end customized data acquisition and delivery services tailored to meet the specific data needs of diverse clients.“Audio-Visual-Motion” Multimodal Synchronized Acquisition (Coming Soon)Synchronized acquisition of multilingual voice commands, human-robot interaction dialogue, and environmental acoustics – bridging the cross‑modal data loop between sound and motion to deliver richer Embodied AI interaction experiences.
+86-21-56472866 / 13310099680
1190 Nong METACINE (South Gate), Building 1, CHINGMU, Zhengtai Road,Baoshan District, Shanghai
Hot searches: Motion capture, Optical motion capture, Motion capture system robot, Motion capture, Bionic robot, Motion control, Gait analysis, Rehabilitation training action assessment, Animation production, Virtual idol live streaming, Expression capture, Virtual production, Pre-visualization, Performance capture
