SimAesthetics

This is a simple ongoing image aesthetic assessment task.

Thanks to the coworker: Kun Xiong. We also thank all the advisers!

Installing

git clone https://github.com/dieuroi/SimAesthetics.git
cd SimAesthetics
virtualenv -p python3.8 env
source ./env/bin/activate

Dataset

The model was trained on the AVA (Aesthetic Visual Analysis) dataset. You can get it from here. The meta data can be found from here.

Models for this task

• NIMA: NIMA: Neural Image Assessment.
• MLSP: Effective Aesthetics Prediction with Multi-Level Spatially Pooled Features.

Pre-trained model (In Progress)

Usage

Usage: cli.py [OPTIONS] COMMAND [ARGS]...

Options:
  --help  Show this message and exit.

Commands:
  get_image_score  Get image scores
  prepare_dataset  Parse, clean and split dataset
  train_model      Train model
  validate_model   Validate model

Contributing

Contributing are welcome

License

This project is licensed under the MIT License - see the LICENSE file for details

Acknowledgments

• Most of this code is borrowed heavily from the Pytorch NIMA PyTorch NIMA: Neural IMage Assessment
• pytorch-mobilenet-v2
• origin NIMA article
• origin MobileNetV2 article
• Post at Google Research Blog
• origin MLSP article
• MLSP (tensorflow)