Xingyuan Releases Cross-Embodiment Data Engine Chorus: Supports Unified Processing of Data from Multiple Devices, Achieving Approximately 40x Training Data Expansion
行元发布机器人跨本体数据引擎Chorus：“不挑设备”，可实现约40倍训练数据扩容
Xingyuan Intelligent released the Cross-Embodiment Data Engine Chorus, which supports converting human operation videos collected from various devices into pre-trained data suitable for different robot embodiments. Testing shows it can achieve approximately 40x effective training data expansion, reducing the equivalent cost per training sample to 5 yuan per hour.
The release highlights a cross-embodiment data engine designed to convert human operation videos into robot training data, with claims of 40x data expansion and 85% data conversion rate. The tool is positioned as a key component in the company's System Scaling approach to embodied intelligence.
Source: Chinese robotics — Research news discovery · Read original article ↗
English translation unavailable. Choose the original article to read it.
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
effective training data expansion
≈40
effective training data expansion
≈40
View original evidence
在当前测试配置下，Chorus 通过 16 种机器人本体适配，将每小时有效机器人轨迹的处理成本降至 15 元；结合 3 倍指令增广，可实现约 40 倍的有效训练数据扩容，单位训练样本的等效成本仅需 5 元/小时。
Open source S12cost per hour of effective robot trajectories
15
View original evidence
在当前测试配置下，Chorus 通过 16 种机器人本体适配，将每小时有效机器人轨迹的处理成本降至 15 元；结合 3 倍指令增广，可实现约 40 倍的有效训练数据扩容，单位训练样本的等效成本仅需 5 元/小时。
Open source S12equivalent cost per training sample
5
View original evidence
在当前测试配置下，Chorus 通过 16 种机器人本体适配，将每小时有效机器人轨迹的处理成本降至 15 元；结合 3 倍指令增广，可实现约 40 倍的有效训练数据扩容，单位训练样本的等效成本仅需 5 元/小时。
Open source S12number of robot embodiments supported
16
View original evidence
在当前测试配置下，Chorus 通过 16 种机器人本体适配，将每小时有效机器人轨迹的处理成本降至 15 元；结合 3 倍指令增广，可实现约 40 倍的有效训练数据扩容，单位训练样本的等效成本仅需 5 元/小时。
Open source S12
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
行元发布机器人跨本体数据引擎Chorus：“不挑设备”，可实现约40倍训练数据扩容
Open source S1
在当前测试配置下，Chorus 通过 16 种机器人本体适配，将每小时有效机器人轨迹的处理成本降至 15 元；结合 3 倍指令增广，可实现约 40 倍的有效训练数据扩容，单位训练样本的等效成本仅需 5 元/小时。
Open source S12
数据处理过程中，Chorus通过三维运动重建、轨迹优化与质量校验，减少相机移动、手部遮挡等因素对转换结果的影响，将平均可用的数据转化率提升至85%。
Open source S11
Source：Chinese robotics — Research news discovery · sohu.com