standoutw commited on
Commit
05aae7b
·
verified ·
1 Parent(s): d476482

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. README.md +155 -3
  2. final_0.zip +3 -0
  3. final_1.zip +3 -0
  4. final_2.zip +3 -0
  5. final_3.zip +3 -0
  6. final_4.zip +3 -0
  7. final_5.zip +3 -0
README.md CHANGED
@@ -1,3 +1,155 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Open Cortex FX v3
2
+
3
+ A curated dataset of videos depicting human manual labor and physical work, organized by task categories.
4
+
5
+ ## Dataset Description
6
+
7
+ Open Cortex FX v3 is a classified video dataset focusing on human manual labor and physical work activities. Each video has been carefully annotated to identify work-related content and categorized into specific labor types.
8
+
9
+ ### Dataset Statistics
10
+
11
+ - **Total Videos**: Classified videos showing human manual labor
12
+ - **Categories**: 20+ distinct labor categories
13
+ - **Format**: MP4 video files organized by category
14
+ - **Structure**: Videos are split into multiple zip files for easy distribution
15
+
16
+ ## Dataset Structure
17
+
18
+ The dataset is organized as follows:
19
+
20
+ ```
21
+ final_0.zip
22
+ final_1.zip
23
+ final_2.zip
24
+ ...
25
+ ```
26
+
27
+ Each zip file contains:
28
+ - **Category folders**: Videos organized by labor type (e.g., `Cooking/`, `Repair/`, `Construction/`)
29
+ - **Metadata CSV**: Mapping file with video information
30
+
31
+ ### Category Folders
32
+
33
+ Videos are organized into category folders based on the type of manual labor depicted:
34
+
35
+ - **Construction** - Building, construction work, using construction tools
36
+ - **Cooking** - Preparing food, cooking, chopping, mixing ingredients
37
+ - **Repair** - Fixing, repairing, maintenance work
38
+ - **Cleaning** - Cleaning, mopping, scrubbing, organizing spaces
39
+ - **Assembly** - Assembling products, putting things together
40
+ - **Gardening** - Gardening, landscaping, outdoor manual work
41
+ - **Painting** - Painting surfaces, applying paint or finishes
42
+ - **Sewing** - Sewing, textile work, fabric manipulation
43
+ - **Woodworking** - Working with wood, carpentry, wood crafting
44
+ - **Metalworking** - Working with metal, welding, metal fabrication
45
+ - **Moving** - Lifting, carrying, moving heavy objects
46
+ - **Serving** - Serving food or drinks, manual service work
47
+ - **Organizing** - Arranging, organizing physical items
48
+ - **Crafting** - General crafting, artisan work, making things by hand
49
+ - **Electrical** - Electrical work, wiring, electrical installation
50
+ - **Plumbing** - Plumbing work, pipe installation, water systems
51
+ - **Automotive** - Automotive repair, working on vehicles
52
+ - **Farming** - Farming, agriculture, working with crops/animals
53
+ - **Welding** - Welding, joining metals with heat
54
+ - **General Labor** - Other manual labor that doesn't fit above categories
55
+
56
+ ## Metadata Format
57
+
58
+ Each zip file includes a `metadata.csv` file with the following columns:
59
+
60
+ | Column | Description |
61
+ |--------|-------------|
62
+ | `new_name` | Randomly assigned filename for the video (e.g., `aBc123XyZ789.mp4`) |
63
+ | `original_name` | Original video filename from source dataset |
64
+ | `category` | Labor category classification |
65
+ | `original_part` | Original dataset part number (for reference) |
66
+ | `caption` | Text description of the video content |
67
+ | `aesthetic_score` | Aesthetic quality score |
68
+ | `motion_score` | Motion activity score |
69
+ | `temporal_consistency` | Temporal consistency score |
70
+ | `camera_motion` | Camera movement type (static, pan_left, zoom_in, etc.) |
71
+ | `frame` | Number of frames in the video |
72
+ | `fps` | Frames per second |
73
+ | `seconds` | Video duration in seconds |
74
+
75
+ ## Usage
76
+
77
+ ### Downloading the Dataset
78
+
79
+ The dataset is available on Hugging Face. You can download it using:
80
+
81
+ ```python
82
+ from huggingface_hub import snapshot_download
83
+
84
+ snapshot_download(
85
+ repo_id="Standout/open-cortex-fx-v3",
86
+ repo_type="dataset",
87
+ local_dir="./open-cortex-fx-v3"
88
+ )
89
+ ```
90
+
91
+ ### Extracting and Using the Data
92
+
93
+ 1. **Extract zip files**: Unzip the downloaded files to access the video content
94
+ 2. **Read metadata**: Load `metadata.csv` from each zip to get video information
95
+ 3. **Access by category**: Navigate to category folders to find specific types of labor videos
96
+
97
+ ### Example: Loading Metadata
98
+
99
+ ```python
100
+ import zipfile
101
+ import pandas as pd
102
+ import io
103
+
104
+ # Extract and read metadata from a zip file
105
+ with zipfile.ZipFile('final_0.zip', 'r') as z:
106
+ with z.open('metadata.csv') as f:
107
+ metadata = pd.read_csv(io.BytesIO(f.read()))
108
+
109
+ # Filter by category
110
+ cooking_videos = metadata[metadata['category'] == 'Cooking']
111
+ print(f"Found {len(cooking_videos)} cooking videos")
112
+ ```
113
+
114
+ ### Example: Accessing Videos
115
+
116
+ ```python
117
+ import zipfile
118
+
119
+ # Extract a specific video
120
+ with zipfile.ZipFile('final_0.zip', 'r') as z:
121
+ # List all videos in Cooking category
122
+ cooking_videos = [name for name in z.namelist() if name.startswith('Cooking/')]
123
+
124
+ # Extract a video
125
+ z.extract('Cooking/example_video.mp4', './output/')
126
+ ```
127
+
128
+ ## Dataset Characteristics
129
+
130
+ - **Focus**: Human manual labor and physical work activities
131
+ - **Quality**: Videos are filtered to show clear human action performing manual tasks
132
+ - **Diversity**: Multiple categories covering various types of physical work
133
+ - **Organization**: Structured by labor type for easy access and filtering
134
+
135
+ ## Citation
136
+
137
+ If you use this dataset in your research, please cite:
138
+
139
+ ```bibtex
140
+ @dataset{open_cortex_fx_v3,
141
+ title={Open Cortex FX v3: A Classified Dataset of Human Manual Labor},
142
+ author={Standout},
143
+ year={2024},
144
+ url={https://huggingface.co/datasets/Standout/open-cortex-fx-v3}
145
+ }
146
+ ```
147
+
148
+ ## License
149
+
150
+ [Specify your license here]
151
+
152
+ ## Contact
153
+
154
+ For questions or issues, please contact [your contact information].
155
+
final_0.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b21bd3918f9516aa81aae5d61ef810f38c45e9850af7f5d291a937fb3afd88c4
3
+ size 418333199
final_1.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0816110a12cca6bc8628bf8bb0e9181770791ff597824b4577a89397c0b3e96
3
+ size 553205201
final_2.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4809aad041a80dcf8f18bf03bf65bfca7aa93b7c0b838aee0771e2e8a22bf223
3
+ size 567849876
final_3.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4b48249d05353c50514823e876ccd9ad6bcaf0a163cb82b03f3bd3ced91c407
3
+ size 704114922
final_4.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23c3f262fe9c4af4223cdd6171b5cbbb7adf3dcf7c1d9ec3b937eb92821c9b7c
3
+ size 425639004
final_5.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:619071e791290ec9b6099590a79f822237fa2c9fb420351215d57287e1a946d5
3
+ size 393134380