Update dataset card to Sample Pack v0.1

README.md

@@ -1,225 +1,178 @@
----
-license: cc-by-nc-4.0
-task_categories:
-  - robotics
-  - reinforcement-learning
-  - depth-estimation
-
-tags:
-  - robotics
-  - manipulation
-  - embodied-ai
-  - imitation-learning
-  - rgbd
-  - realsense
-  - kitchen
-  - cooking
-  - human-demonstration
----
-
-# 🍳 Chinese Commercial Kitchen Manipulation Dataset — Preview Pack
-
-> Asia's first real commercial kitchen manipulation dataset.
->
-> Professional chef (20 years) · Real restaurant · RGB-D · Multi-view
-
----
-
-# Overview
-
-This repository contains a preview pack of a real-world Chinese commercial kitchen manipulation dataset.
-
-The dataset was collected in an operating restaurant environment using multi-view RGB and depth sensors. Unlike laboratory datasets, the recordings capture realistic cooking workflows, cluttered workspaces, variable lighting conditions, and natural human motion.
-
-The current preview release contains screenshot samples only.
-
-The complete dataset, including original videos, raw depth files, and future task collections, is available upon request.
-
----
-
-# Key Features
-
-- Real commercial Chinese kitchen environment
-- Professional chef with 20 years of experience
-- Multi-view recording setup
-- RGB + Depth data
-- First-person (egocentric) perspective
-- Suitable for:
-  - Embodied AI
-  - Robotic Manipulation
-  - Imitation Learning
-  - Human Action Recognition
-  - Visual-Language Action Models
-  - Grasp Planning Research
-
----
-
-# Preview Contents
-
-## RGB — Overhead View (`/overhead`)
-
-Fixed Intel RealSense D435I camera mounted approximately 1.5 meters above the workstation.
-
-Provides a complete view of:
-
-- Ingredients
-- Tools
-- Both hands
-- Cooking workspace
-
----
-
-## Depth Data — Colorized Preview (`/depth`)
-
-Raw depth data was captured using an Intel RealSense D435I sensor.
-
-Preview images are colorized visualizations generated from the original depth frames.
-
-Depth format:
-
-- HDF5
-- Float32
-- Metric depth (meters)
-
-Point cloud reconstruction can be generated directly from the original depth recordings.
-
----
-
-## Task 1 — Cutting (`/task_01_cutting`)
-
-Examples include:
-
-- Vegetable cutting
-- Meat cutting
-
-Views:
-
-- Egocentric (head-mounted camera)
-- Side view camera
-- Overhead RGB-D (Intel RealSense D435I, includes depth data)
-
----
-
-## Task 2 — Wok Stir-Fry (`/task_02_stir_fry`)
-
-High-difficulty bimanual manipulation task involving:
-
-- Ingredient handling
-- Wok operation
-- Continuous tool interaction
-
-Views:
-
-- Egocentric (head-mounted camera)
-- Side view camera
-
----
-
-# Sensor Configuration
-
-Three complementary viewpoints were recorded simultaneously:
-
-| View | Resolution | Frame Rate |
-|--------|--------|--------|
-| Egocentric | 3840 × 2160 (4K) | 29.97 fps |
-| Side View | 1920 × 1080 | 60 fps |
-| Overhead RGB-D | 1280 × 720 | 15 fps |
-
-This multi-view setup captures both fine-grained hand-object interactions and global workspace context.
-
----
-
-# Available Data
-
-| Content | Format | Status |
-|----------|----------|----------|
-| Multi-view RGB Videos | MP4 | Available upon request |
-| Depth Data | HDF5 (float32, meters) | Available upon request |
-| RealSense Raw Recordings | .bag | Available upon request |
-| Screenshot Preview Pack | PNG | Included in this repository |
-
----
-
-# Future Collection Tasks
-
-Additional task categories are currently being planned and can be collected upon request.
-
-Potential tasks include:
-
-- Stewing
-- Deep Frying
-- Pan Frying
-- Dumpling Folding
-- Ingredient Preparation
-- Kitchen Cleaning Procedures
-
-Custom task collection may be available for research and commercial projects.
-
----
-
-# Collection Environment
-
-**Location**
-
-Zhongshan, Guangdong, China
-
-**Venue**
-
-Operating commercial Chinese restaurant
-
-**Operator**
-
-Professional chef with approximately 20 years of experience
-
-**Consent**
-
-Full informed consent obtained from all participants.
-
----
-
-# Dataset Applications
-
-Potential applications include:
-
-- Robotic Cooking Systems
-- Embodied Foundation Models
-- Visual Action Understanding
-- Human Demonstration Learning
-- Multi-modal Perception
-- RGB-D Manipulation Research
-- Human-Robot Collaboration
-
----
-
-# Access to Full Dataset
-
-The complete dataset is not publicly hosted due to storage size limitations.
-
-Available materials include:
-
-- Multi-view MP4 recordings
-- Original RealSense depth data
-- Raw bag recordings
-- Optional annotations
-
-Researchers and organizations interested in accessing the complete dataset may contact:
-
-📧 andy@dynamicnova.com
-
-Please include:
-
-- Research topic
-- Intended use case
-- Preferred format (MP4 / HDF5 / RLDS / LeRobot)
-
----
-
-# Citation
-
-If you use this dataset in academic or commercial research, please cite this repository.
-
----
-
-*Preview collected in May 2026*
-
-*Nova Dynamics Limited*
+---
+license: cc-by-nc-4.0
+task_categories:
+  - robotics
+  - reinforcement-learning
+  - depth-estimation
+
+tags:
+  - robotics
+  - manipulation
+  - embodied-ai
+  - imitation-learning
+  - rgbd
+  - realsense
+  - kitchen
+  - cooking
+  - human-demonstration
+---
+
+# 🍳 Chinese Commercial Kitchen Manipulation Dataset — Sample Pack v0.1
+
+> **Asia's first real commercial kitchen manipulation dataset.**  
+> Professional chef (20 years) · Real restaurant environment · Multi-view RGB-D · Egocentric video
+
+**📧 Request evaluation samples or full data:** [andy@dynamicnova.com](mailto:andy@dynamicnova.com)
+
+---
+
+## Overview
+
+This sample pack contains real-world cooking demonstrations collected in an operating Chinese commercial kitchen in Zhongshan, Guangdong, China.
+
+The data focuses on professional chef workflows rather than staged tabletop demonstrations. It includes synchronized or task-aligned multi-view video, egocentric footage, and metric depth data for evaluating robotics, embodied AI, imitation learning, and visual action understanding pipelines.
+
+Related work such as EgoMimic suggests that egocentric human demonstration data can be valuable for scaling imitation learning, especially when paired with robot data or aligned sensing setups.
+
+Key characteristics:
+
+- Real commercial Chinese restaurant kitchen
+- Professional chef with approximately 20 years of experience
+- Egocentric, side-view, and overhead camera perspectives
+- Intel RealSense D435I RGB-D capture for overhead view
+- Chinese cooking tasks involving tool use, bimanual coordination, and fine-grained food manipulation
+
+---
+
+## Sample Pack Contents
+
+### Task 1 — Cutting Vegetables (切菜) · 3 camera views + depth
+
+| File | Description |
+|------|-------------|
+| `egocentric.mp4` | Head-mounted action camera, first-person view (3840×2160, 29.97fps) |
+| `side_view.mp4` | Fixed side-view phone camera (1920×1080, 60fps) |
+| `overhead/overhead.mp4` | Fixed overhead RealSense RGB (1280×720, 15fps) |
+| `depth/depth.hdf5` | Aligned depth frames, float32 in meters (480×848, 2379 frames) |
+
+### Task 2 — Wok Stir-Fry (翻炒) · 2 camera views
+
+| File | Description |
+|------|-------------|
+| `egocentric.mp4` | Head-mounted action camera, first-person view (3840×2160, 29.97fps) |
+| `side_view.mp4` | Fixed side-view phone camera (1920×1080, ~60fps) |
+
+> Depth sample is included for Task 1. Additional depth recordings may be available depending on the task and capture setup.
+
+> *Depth note: Pixels with value 65.535m indicate no valid depth return (sensor limit). Typical valid pixel rate: ~86%.*
+
+> Preview PNG/JPG images are included alongside the videos for quick browsing.
+
+---
+
+## File Structure
+
+```
+samplepack_video/
+├── task_01_cutting/
+│   ├── egocentric.mp4
+│   ├── side_view.mp4
+│   ├── overhead/
+│   │   ├── overhead.mp4
+│   │   └── overhead_*.PNG
+│   ├── depth/
+│   │   ├── depth.hdf5
+│   │   └── check_*.jpg
+│   ├── egocentric_screenshot_*.PNG
+│   └── side_view_*.PNG
+└── task_02_stir_fry/
+    ├── egocentric.mp4
+    ├── side_view.mp4
+    ├── egocentric_screenshot_*.PNG
+    └── side_view_*.PNG
+```
+
+**Read depth data (Python):**
+```python
+import h5py
+with h5py.File("samplepack_video/task_01_cutting/depth/depth.hdf5", "r") as f:
+    depth = f["depth_meters"][:]   # (2379, 480, 848) float32, meters
+    ts    = f["timestamps"][:]
+```
+
+---
+
+## Camera Setup
+
+| View | Device | Resolution | Frame Rate |
+|------|--------|------------|------------|
+| Egocentric | Head-mounted action camera | 3840×2160 (4K) | 29.97 fps |
+| Side | Fixed smartphone | 1920×1080 | 60 fps |
+| Overhead RGB-D | Intel RealSense D435I | RGB 1280×720 / Depth 848×480 | 15 fps |
+
+---
+
+## Collection Environment
+
+- **Location:** Zhongshan, Guangdong, China
+- **Venue:** Operating commercial Chinese restaurant
+- **Operator:** Professional chef, 20 years experience
+- **Consent:** Full informed consent obtained from all participants
+
+---
+
+## Tasks
+
+| Task | Chinese | Difficulty | Bimanual | Camera Views | Depth |
+|------|---------|-----------|----------|-------------|-------|
+| Cutting vegetables | 切菜 | Medium | Partial | 3 | ✅ |
+| Wok stir-fry | 翻炒 | High | ✅ | 2 | Available upon request |
+
+**Planned in full dataset:**
+
+| Task | Chinese | Key Challenge |
+|------|---------|--------------|
+| Dumpling folding | 包饺子 | High dexterity, bimanual |
+| Dough kneading | 揉面 | Force estimation, rhythm |
+| Deep frying | 炸 | Temperature judgment, timing |
+| Pan frying | 煎 | Heat control, single/double-side flip |
+| Braising / stewing | 炖煮 | Long-horizon, multi-step sequencing |
+| Sauce thickening | 勾芡 | Fine motor control, timing-sensitive |
+| Marinating / seasoning | 腌制/调味 | Multi-ingredient coordination |
+
+*Action-level labels (e.g. flip timing, thickening moment, seasoning sequence) available upon request.*
+
+---
+
+## Access to Full Data
+
+This public sample pack is intended for technical evaluation and early research feedback.
+
+Additional materials may be available upon request:
+
+- Longer multi-view MP4 recordings
+- Additional HDF5 metric depth sequences
+- RealSense raw `.bag` recordings
+- Task-level or action-level annotations
+- Format conversion support, including HDF5, RLDS, or LeRobot
+- Custom collection for specific kitchen workflows
+
+---
+
+## Contact
+
+📧 **andy@dynamicnova.com**
+
+Please include: tasks of interest, required volume, preferred format (HDF5 / RLDS / LeRobot), timeline.
+
+Response within 48 hours.
+
+---
+
+## Citation
+
+If you use this dataset in academic or commercial research, please cite this repository.
+
+---
+
+*Collected May 2026 · Zhongshan, Guangdong, China · Nova Dynamics Limited*