---
license: cc
task_categories:
- image-to-image
tags:
- emojis
- superresolution
pretty_name: Emoji LIIF
size_categories:
- 1K<n<10K
---
# Dataset Card for emoji-liif

This dataset contains 3,165 high-resolution emoji images that have been upscaled using the LIIF (Learning Implicit Image Function) method. The images were generated from original Apple emoji assets and are provided for research and academic purposes under fair use.

## Dataset Details

### Dataset Description

The emoji-liif dataset consists of upscaled emoji images generated from Apple's emoji assets. Each image has been enlarged to 2000x2000 pixels using the LIIF method, which provides high-quality upscaling while preserving image details.

- **Curated by:** Louis Maddox (lmmx)
- **Language(s) (NLP):** Not applicable - this is an image dataset
- **License:** Research/Academic Use Only (Fair Use)

### Dataset Sources

- **Repository:** https://github.com/lmmx/emoji-liif
- **Paper [optional]:** Original LIIF paper: ["Learning Continuous Image Representation with Local Implicit Image Function"](https://arxiv.org/abs/2012.09161)

## Uses

### Direct Use

This dataset is intended for:
- Research on image super-resolution techniques
- Comparative analysis of upscaling methods
- Academic study of emoji representation and scaling
- Benchmark dataset for super-resolution algorithms

### Out-of-Scope Use

This dataset should not be used for:
- Commercial purposes
- Distribution of emoji images as part of products or services
- Any use that violates Apple's copyright on the original emoji designs

## Dataset Structure

The dataset consists of 3,165 PNG image files, each with the following characteristics:
- Resolution: 2000x2000 pixels
- Format: PNG
- Organization: Single directory containing all images
- Source: Upscaled from Apple emoji assets (iOS 14.2 72x72 and OSX 10.15 160x160)

## Dataset Creation

### Curation Rationale

The dataset was created to demonstrate the capabilities of the LIIF upscaling method on emoji images, providing high-quality enlarged versions while preserving the original design details.

### Source Data

#### Data Collection and Processing

The images were generated through the following process:
1. Original emoji assets were extracted from Apple's iOS 14.2 and OSX 10.15 systems
2. Images were upscaled using the LIIF method
3. Final images were saved as 2000x2000 pixel PNGs

#### Who are the source data producers?

The original emoji designs are created by Apple Inc. The upscaled versions were produced using the LIIF method by the repository maintainer.

### Personal and Sensitive Information

This dataset does not contain any personal or sensitive information. The images are emoji characters that are publicly available as part of operating systems.

## Bias, Risks, and Limitations

- The dataset only includes Apple emoji designs and may not be representative of other emoji styles
- The upscaling process may introduce artifacts or slight modifications to the original designs
- The dataset is limited to fair use applications due to copyright considerations

### Recommendations

Users should:
- Respect the fair use limitations of the dataset
- Acknowledge Apple's copyright of the original emoji designs
- Cite both the dataset and the LIIF paper when using this data in research

## Citation

If you use this dataset, please cite:

```bibtex
@misc{emoji-liif,
  author = {Maddox, Louis},
  title = {emoji-liif: Emoji upsampling workflow using LIIF},
  year = {2024},
  publisher = {GitHub},
  url = {https://github.com/lmmx/emoji-liif}
}
```

## Dataset Card Contact

For questions about this dataset, please open an issue on the GitHub repository: https://github.com/lmmx/emoji-liif/issues