📊
Dataset

Humanoid Everyday Stepit

by UsanoCoCr usanococr/humanoid-everyday-stepit
Free2AITools Nexus Index
61.1
S: Semantic 50

Query-time baseline · scored live at search

A: Authority 62
P: Popularity 56
R: Recency 97
Q: Quality 50
Tech Context
Vital Performance
Data Integrity 61.1 FNI Score
- Size
- Rows
- Tokens
Dataset Information Summary
Entity Passport
Registry ID usanococr/humanoid-everyday-stepit
Provider huggingface
📜

Cite this dataset

Academic & Research Attribution

BibTeX
@misc{hf_dataset_usanococr_humanoid_everyday_stepit,
  author = {UsanoCoCr},
  title = {Humanoid Everyday Stepit Dataset},
  year = {2026},
  howpublished = {\url{https://huggingface.co/datasets/UsanoCoCr/humanoid-everyday-stepit}},
  note = {Accessed via Free2AITools.}
}
APA Style
UsanoCoCr. (2026). Humanoid Everyday Stepit [Dataset]. Free2AITools. https://huggingface.co/datasets/UsanoCoCr/humanoid-everyday-stepit

🔬Technical Deep Dive

Full Specifications [+]

⚖️ Free2AITools Nexus Index V2.0

Semantic (S) 50

Query-time baseline · scored live at search

Authority (A) 62
Popularity (P) 56
Recency (R) 97
Quality (Q) 50

💬 Index Insight

FNI V2.0 for Humanoid Everyday Stepit: Authority (A:62), Popularity (P:56), Recency (R:97), Quality (Q:50). Semantic (S) is a query-time baseline scored live at search.

Free2AITools Nexus Index

Data Sources / Provenance

Open data Updated: Live data
⬇️
Downloads
67,376

👁️ Data Preview

📊

Row-level preview not available for this dataset.

Schema structure is shown in the Field Logic panel when available.

🔗 Explore Full Dataset ↗

🧬 Field Logic

🧬

Schema not yet indexed for this dataset.

Dataset Specification

Humanoid-Everyday · stepit action(G1 子集)

一个自包含的标准 LeRobot v2.1 数据集,由原始 Humanoid-Everyday 数据集的 Unitree G1 子集 重新表达而来。与原始数据唯一的区别是 action 列被替换成了一个 41 维、包含完整 base 状态 (位置 + 姿态 + 线速度)的动作向量;其余所有列均逐字节保持不变,可直接用标准 LeRobot 加载器读取。

  • 动作空间:41 维 = base(10) + G1 身体 29 关节 + 双手开合 2;抽取相应切片即可喂给 stepit(Unitree G1 29-DOF)控制器(见下文 stepit qpos)。
  • 规模:4068 episodes / 1,781,092 frames / 246 tasks / 9 chunks,fps = 30robot_type = g1
  • 体积:≈ 394 GB(parquet ≈ 391 GB + 视频 ≈ 3 GB + meta ≈ 2 MB)。

原始数据是 G1/H1 混合的(共 8954 episodes)。stepit 只支持 Unitree G1,因此 4886 个 H1 episode 被跳过,只保留 4068 个 G1 episode(被跳过的清单见 meta/skipped_manifest.json)。


目录结构

text
data_stepit/
├── data/                                 # 每个 episode 一个 parquet(逐帧 observation + action)
│   └── chunk-XXX/episode_YYYYYY.parquet
├── videos/                               # 每个 episode 一个 mp4(第一视角 RGB,H.264)
│   └── chunk-XXX/egocentric/episode_YYYYYY.mp4
└── meta/
    ├── info.json                         # 数据集级元数据 + features schema(action 为 41 维)
    ├── stats.json                        # 全局逐列统计(action + 若干 observation 列)
    ├── episodes.jsonl                    # 每 episode 的 length / tasks / 全局帧区间 / instruction
    ├── episodes_stats.jsonl              # 每 episode 逐列统计(占位,保留 LeRobot 结构)
    ├── tasks.jsonl                       # task_index → 任务名(共 246 个 task)
    ├── hand_thresholds.json              # 手部开/合二值化所用的 Otsu 阈值及标定信息
    ├── converted_manifest.json           # 已转换 episode 清单
    └── skipped_manifest.json             # 被跳过的非 G1 episode 清单(含关节维度 signature)
  • 文件名沿用原始稀疏编号episode_index 不连续),parquet 与 mp4 通过 episode_index 一一对齐。
  • 编号约定:chunk = episode_index // 1000;路径模板见 info.jsondata_path / video_path

坐标系与单位约定

单位 坐标系
位置 position 米 (m) 里程计/世界系
四元数 quat 无量纲,顺序 (w, x, y, z) 里程计/世界系
线速度 velocity / base_lin_vel m/s base(机体)系(已实测确认,见下文)
角速度 gyroscope rad/s base 系
关节角 弧度 (rad)

Social Proof

HuggingFace Hub
67.4KDownloads
🔄 Updated daily

Source summary: Based on Hugging Face metadata. Not a recommendation.

📊 FNI Methodology 📚 Knowledge Baseℹ️ Verify with original source

🛡️ Dataset Transparency Report

Technical metadata sourced from upstream repositories.

Open Metadata

🆔 Identity & Source

id
hf-dataset--usanococr--humanoid-everyday-stepit
slug
usanococr--humanoid-everyday-stepit
source
huggingface
author
UsanoCoCr
license
tags
size_categories:1m<n<10m, format:parquet, modality:tabular, modality:timeseries, modality:video, library:datasets, library:dask, library:polars, library:mlcroissant, region:us

⚙️ Technical Specs

architecture
null
params billions
null
context length
null
pipeline tag

📊 Engagement & Metrics

downloads
67,376
stars
null
forks
null

Data indexed from public sources. Updated daily.