|
--- |
|
license: apache-2.0 |
|
task_categories: |
|
- text-to-image |
|
size_categories: |
|
- 1K<n<10K |
|
configs: |
|
- config_name: default |
|
data_files: data/**/* |
|
language: |
|
- en |
|
tags: |
|
- finetune |
|
- overlai.ai |
|
- c2pa |
|
- contentcredentials |
|
- opt-in |
|
- polar |
|
- bear |
|
- subarctic |
|
- polarbear |
|
- arctic |
|
- wildlife |
|
--- |
|
 |
|
# Sub Arctic Polar Bear |
|
|
|
<!-- Provide a quick summary of the dataset. --> |
|
|
|
**"Sub Arctic Polar Bear"** is a fine tuning text-to-image dataset consisting of dynamic photographs captured in 30MP+ resolution using Canon R5 and a Canon 600mm RF lens. |
|
|
|
## Key Features |
|
|
|
- 📷 **Detail**: Every photograph is captured at the proper exposure, full sharpness, delivering rich detail ideal for fine tuning polar bear datasets. |
|
- 🎞️ **Variance**: Includes close-up details, behavioural, low perspective ground level, all angles of the polar bears, medium and far shots in habitat, autumn and winter scenes. |
|
- ⚙️ **Consistency**: Re-thinking training data at the point of capture by "overshooting" a subject, enabling models to learn more nuanced relationships and views across scenes. |
|
- 🌅 **Light**: Shot during early morning, sunset light for optimal color contrast and dynamic range. Also, shot during bright overcast, maximizing visual quality for color and lighting-sensitive tasks. |
|
- 🔍 **Curation**: Curated specifically for machine learning, providing clean, high-quality data for next generation model training. |
|
|
|
## Dataset Details |
|
|
|
- **Total photos:** 1,801 |
|
- **Total size:** 55.01 GB |
|
- **Photo types:** Close-ups, medium distance, habitat, portraits, side profile, all angles, awake, walking, sleeping, eating. |
|
|
|
## Technical Details |
|
|
|
- **Cameras:** Canon R5 |
|
- **Lenses:** Canon 600mm RF, Canon 70-200mm EF |
|
- **Resolution:** 8192x5464 |
|
- **Colorspace:** Adobe RGB 1998 |
|
- **Image Size:** 35MP+ |
|
|
|
### Content Credentials |
|
|
|
Each video in the "Sub Arctic Polar Bear" dataset contains **C2PA Content Credentials** metadata and can be used with the [Content Credentials Space by Truepic](https://huggingface.co/spaces/Truepic/ai-content-credentials) to begin testing model provenance. |
|
|
|
## What is Overlai.ai? |
|
|
|
[Overlai.ai](https://overlai.ai/) specializes in 8K+ fine tuning datasets with the goal of improving photo & video models. Contact us for the complete dataset list or access to this dataset in full resolution. |
|
|
|
## Contact |
|
|
|
*[email protected]* |
