refine-axis/axis-v1_1k

---
license: cc-by-4.0
task_categories:
- image-to-text
- text-to-image
language:
- en
tags:
- art
- photography
- ai-training
- captions
size_categories:
- 1K<n<10K
---

# 🚀 Axis-V1 Image Dataset

Welcome to the official repository of **Axis-V1**, a high-quality curated dataset designed for advanced AI training, specifically for Vision-Language Models (VLM) and Image Generation tasks. Developed by **refine-axis**, this dataset provides a diverse range of visual concepts paired with dense, professional descriptive captions.

## 📊 Dataset Overview

The dataset contains a total of **1,000 high-resolution images**, each accompanied by a corresponding `.txt` file containing a detailed descriptive paragraph focusing on textures, lighting, object anatomy, and composition.

### 📂 Folder Structure & Statistics

The dataset is organized into **8 specialized sub-folders**:

| # | Folder Name | Files Count (Images + Text) | Description |
|---|-------------|-----------------------------|-------------|
| 1 | `Animals_Photo` | 100 + 100 | Wildlife and pets with focus on fur texture and anatomy. |
| 2 | `Architecture_&_Interiors Photo` | 100 + 100 | Urban structures, modern minimalist, and classic interior designs. |
| 3 | `Cinematic_Photo` | 100 + 100 | Photos with dramatic lighting, deep shadows, and storytelling vibes. |
| 4 | `Human_Photo` | 100 + 100 | Portraits and lifestyle shots focusing on skin tones and expressions. |
| 5 | `Nature_&_Macro Texture` | 175 + 175 | Extreme close-ups of natural elements and intricate textures. |
| 6 | `Night_&_Neon Photo` | 175 + 175 | Low-light photography with vibrant neon colors and light trails. |
| 7 | `Robotics_Photo` | 100 + 100 | Futuristic designs, metallic surfaces, and mechanical details. |
| 8 | `Vehicles_Photo` | 150 + 150 | Automotive photography covering classic cars to modern vehicles. |

**Total Data: 1,000 Images & 1,000 Descriptive Labels.**

## 📊 Image Quality Benchmark

Axis-V1 is not just a collection of images; it is a high-fidelity dataset engineered for precision. We conducted a rigorous automated audit using **Laplacian Variance** (Sharpness) and **Pixel-Density Mapping** (Resolution) across all 1,000 images.

| Category | Sharpness Score | Avg Resolution | Status |
| :--- | :---: | :---: | :--- |
| Animals_Photo | 445.79 | 17.94 MP | ✅ Cinematic Focus |
| Architecture_&_Interiors Photo | 519.38 | 23.01 MP | ✨ High Quality |
| Cinematic_Photo | 507.41 | 24.98 MP | ✨ High Quality |
| Human_Photo | 325.01 | 20.14 MP | ✅ Cinematic Focus |
| Nature_&_Macro Texture | 1007.65 | 17.66 MP | 💎 Premium Sharp |
| Night_&_Neon Photo | 311.22 | 19.85 MP | ✅ Cinematic Focus |
| Robotics_Photo | 298.22 | 20.74 MP | ✅ Cinematic Focus |
| Vehicles_Photo | 248.77 | 20.55 MP | ✅ Cinematic Focus |

## 🎯 Purpose of Dataset

This dataset was built using an automated pipeline with human-in-the-loop oversight to ensure:
- **High Descriptive Quality:** Every image is described for AI training (captions include lighting conditions and material details).
- **Diversity:** Covers everything from nature to futuristic robotics.
- **Ease of Use:** Standardized folder structure for easy loading in PyTorch, TensorFlow, or Hugging Face Transformers.

## 📜 Licensing

This dataset is licensed under **Creative Commons Attribution 4.0 International (CC BY 4.0)**.
You are free to:
- **Share:** Copy and redistribute the material in any medium or format.
- **Adapt:** Remix, transform, and build upon the material for any purpose, even commercially.

*Note: You must give appropriate credit to **refine-axis** as the creator.*

---
**Maintained by:** [refine-axis]
**Release Date:** April 2026