---
license: cc-by-4.0
task_categories:
  - video-classification
  - other
language:
  - en
tags:
  - egocentric
  - first-person-video
  - action-recognition
  - skilled-trades
  - commercial-work
  - construction
  - electrician
  - carpentry
  - fabrication
  - vocational
  - industrial
  - video
pretty_name: Skilled Commercial Work Egocentric Video Dataset Sample
size_categories:
  - n<1K
configs:
  - config_name: default
    data_files:
      - split: train
        path: metadata.csv
---

# 🛠️ Skilled Commercial Work — Egocentric Video Dataset (Sample)

**This dataset is part of a larger collection of egocentric activity datasets by Verbose Tech Labs LLP. If you want the full dataset, or want access to more categories? Get in touch with us:**

- 📞 **Phone:** [+91 7672 000 500](tel:+917672000500)
- 💬 **WhatsApp:** [+91 7672 000 500](https://wa.me/917672000500)
- 📧 **Email:** [Hello@VerboseTechLabs.com](mailto:Hello@VerboseTechLabs.com)
- 🌐 **Website:** [VerboseTechLabs.com](https://VerboseTechLabs.com)
- 🔗 **More datasets:** [kaggle.com/verbosetechlabsllp](https://www.kaggle.com/verbosetechlabsllp)

---

## Dataset Summary

First-person point-of-view (POV) video recordings of skilled commercial work across multiple trades — brick work, electrical work, metal fabrication, frame manufacturing, furniture assembly, and carpentry. This is a **sample release** showcasing the quality and variety of our larger skilled trades dataset collection.

## Dataset Statistics

| Metric | Value |
|---|---|
| Total clips | 8 |
| Total duration | ~100 minutes (~1 hour 40 minutes) |
| Total size | ~4.7 GB |
| Trade categories | 6 |
| View type | Egocentric (first-person) |
| Video format | MP4 |
| Frame rate | 30 fps |
| Resolution | 1080p |

## Supported Tasks

- **Video classification** — classify skilled trade activities
- **Action recognition** — recognize commercial work actions
- **Fine-grained skill classification** — identify specific trade techniques
- **Worker training** — AI-powered tutorial systems
- **Ergonomics research** — repetitive motion analysis
- **Assistive robotics** — training robots for construction & workshops
- **Safety monitoring** — detect risky worker actions
- **Vocational education** — video-based curriculum content

## Dataset Structure

### Folder Structure

```
skilled-commercial-work-egocentric-sample/
├── videos/
│   ├── brick_work/
│   │   └── brick_work_01.mp4
│   ├── electrician/
│   │   ├── electrician_01.mp4
│   │   └── electrician_02.mp4
│   ├── fabrication/
│   │   ├── fabrication_01.mp4
│   │   └── fabrication_02.mp4
│   ├── frames_manufacturing/
│   │   └── frames_manufacturing_01.mp4
│   ├── furniture_assembly/
│   │   └── furniture_assembly_01.mp4
│   └── wood_work/
│       └── wood_work_01.mp4
├── metadata.csv
└── README.md
```

### Data Fields

The `metadata.csv` file contains the following columns:

| Column | Type | Description |
|---|---|---|
| `file_name` | string | Relative path to the video file |
| `clip_id` | string | Unique identifier (e.g., `SCW_001`) |
| `activity` | string | Main trade category |
| `sub_activity` | string | Fine-grained label |
| `duration` | string | Human-readable duration (HH:MM:SS) |
| `duration_seconds` | integer | Duration in seconds |
| `file_size_mb` | float | File size in megabytes |
| `recording_date` | date | Recording date (YYYY-MM-DD) |
| `resolution` | string | Video resolution |
| `fps` | integer | Frames per second |
| `view_type` | string | Camera view type (`egocentric`) |
| `notes` | string | Additional context |

### Activity Breakdown

| Clip ID | File | Trade | Sub-activity | Duration |
|---|---|---|---|---|
| SCW_001 | brick_work_01.mp4 | brick_work | masonry_construction | 00:09:00 |
| SCW_002 | electrician_01.mp4 | electrician | electrical_wiring | 00:42:10 |
| SCW_003 | electrician_02.mp4 | electrician | electrical_wiring | 00:16:34 |
| SCW_004 | fabrication_01.mp4 | fabrication | metal_fabrication | 00:00:30 |
| SCW_005 | fabrication_02.mp4 | fabrication | metal_fabrication | 00:00:30 |
| SCW_006 | frames_manufacturing_01.mp4 | frames_manufacturing | frame_assembly | 00:01:31 |
| SCW_007 | furniture_assembly_01.mp4 | furniture_assembly | furniture_making | 00:15:03 |
| SCW_008 | wood_work_01.mp4 | wood_work | carpentry | 00:14:16 |

### Trade Categories

- 🧱 **Brick Work** — Masonry construction, bricklaying, mortar application
- ⚡ **Electrician** — Electrical wiring, installation, and repair work
- 🔩 **Fabrication** — Metal fabrication, welding preparation
- 🪟 **Frames Manufacturing** — Window/door frame assembly
- 🪑 **Furniture Assembly** — Furniture making, joinery, assembly
- 🪵 **Wood Work** — Carpentry and woodworking shop activities

### Duration Breakdown by Trade

- **Electrician** — 2 clips • ~58 min (longest coverage — extensive wiring work)
- **Wood Work** — 1 clip • ~14 min (carpentry shop)
- **Furniture Assembly** — 1 clip • ~15 min (joinery work)
- **Brick Work** — 1 clip • ~9 min (masonry)
- **Frames Manufacturing** — 1 clip • ~1.5 min (sample)
- **Fabrication** — 2 clips • ~1 min (short samples)

## Usage

### Load with 🤗 datasets library

```python
from datasets import load_dataset

dataset = load_dataset("verbosetechlabsllp/skilled-commercial-work-egocentric-sample")
print(dataset)
```

### Load metadata directly with Pandas

```python
import pandas as pd

df = pd.read_csv("hf://datasets/verbosetechlabsllp/skilled-commercial-work-egocentric-sample/metadata.csv")
print(df.head())
print(df['activity'].value_counts())
```

### Download a specific video

```python
from huggingface_hub import hf_hub_download

video_path = hf_hub_download(
    repo_id="verbosetechlabsllp/skilled-commercial-work-egocentric-sample",
    filename="videos/electrician/electrician_01.mp4",
    repo_type="dataset"
)
print(f"Video downloaded to: {video_path}")
```

### Extract sample frames

```python
import cv2, os

def extract_frames(video_path, out_dir, every_n_seconds=10):
    os.makedirs(out_dir, exist_ok=True)
    cap = cv2.VideoCapture(video_path)
    fps = cap.get(cv2.CAP_PROP_FPS)
    frame_interval = int(fps * every_n_seconds)
    count, saved = 0, 0
    while True:
        ret, frame = cap.read()
        if not ret: break
        if count % frame_interval == 0:
            cv2.imwrite(f"{out_dir}/frame_{saved:04d}.jpg", frame)
            saved += 1
        count += 1
    cap.release()
    return saved
```

## Data Collection

- **Camera view**: First-person / egocentric (head-mounted or chest-mounted)
- **Environment**: Real construction sites, workshops, and commercial work settings
- **Lighting**: Mixed natural and workshop lighting
- **Audio**: Included in MP4 (includes ambient tool and machine sounds — usable for multimodal research)
- **Recording period**: May 2026 – June 2026

## Licensing Information

**CC BY 4.0** — Free for research and commercial use with attribution.

## Citation

```bibtex
@dataset{skilled_commercial_work_egocentric_2026,
  title  = {Skilled Commercial Work — Egocentric Video Dataset (Sample)},
  author = {Verbose Tech Labs LLP},
  year   = {2026},
  url    = {https://huggingface.co/datasets/verbosetechlabsllp/skilled-commercial-work-egocentric-sample}
}
```

## More Datasets from Verbose Tech Labs

This dataset is part of a larger collection of egocentric activity datasets covering:
- 👕 Clothing industry manufacturing
- 🍳 Cooking & food preparation
- 🧹 Household cleaning tasks
- 🏭 Manufacturing unit workflows (sample)
- 🛠️ Skilled commercial work (this — sample)
- ...and more categories in development

🔗 Browse all our datasets: [kaggle.com/verbosetechlabsllp](https://www.kaggle.com/verbosetechlabsllp) | [huggingface.co/verbosetechlabsllp](https://huggingface.co/verbosetechlabsllp)