https://github.com/yiling-chen/view-finding-network

A deep ranking network that learns to find good compositions in a photograph.
https://github.com/yiling-chen/view-finding-network

aesthetics dataset deep-learning evaluation tensorflow

Last synced: 6 days ago
JSON representation

A deep ranking network that learns to find good compositions in a photograph.

• Host: GitHub
• URL: https://github.com/yiling-chen/view-finding-network
• Owner: yiling-chen
• License: gpl-3.0
• Created: 2017-02-03T18:29:53.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2021-11-24T08:42:58.000Z (almost 3 years ago)
• Last Synced: 2024-08-01T15:30:57.421Z (3 months ago)
• Topics: aesthetics, dataset, deep-learning, evaluation, tensorflow
• Language: Python
• Homepage:
• Size: 42.2 MB
• Stars: 74
• Watchers: 5
• Forks: 16
• Open Issues: 7
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# View Finding Network

This repository contains the dataset and scripts used in the following article:

[Yi-Ling Chen](https://yiling-chen.github.io/), Jan Klopp, [Min Sun](http://aliensunmin.github.io/), [Shao-Yi Chien](http://www.ee.ntu.edu.tw/profile?id=101), [Kwan-Liu Ma](http://www.cs.ucdavis.edu/~ma/), ["Learning to Compose with Professional Photographs on the Web"](https://arxiv.org/abs/1702.00503), in Proc. of ACM Multimedia 2017. [(Supplemetnal)](pdf/mm2017-supplemental.pdf)

**News**  Check out this [PyTorch implementation](https://github.com/remorsecs/pytorch-view-finding-network) if you are interested.

## Dependencies

You will need to have `tensorflow` (version > 1.0), `skimage`, `tabulate`, `pillow` installed on your system to run the scripts.

## Download the dataset

* Clone the repository to your local disk.

* Under a command line window, run the following command to get the training images from Flickr:

```bash

$ python download_images.py -w 4

```

The above command will launch 4 worker threads to download the images to a default folder (./images).

## Training

* Run `create_dbs.py` to generate the TFRecords files used by Tensorflow.

* Run `vfn_train.py` to start training.

```bash

$ python vfn_train.py --spp 0

```

The above example starts training with SPP disabled. Or you may want to enable SPP with either `max` or `avg` options.

```bash

$ python vfn_train.py --pooling max

```

Note that if you changed the output filenames when running `create_dbs.py`, you will need to provide the new filenames to `vfn_train.py`. Take a look at the script to check out other available parameters or run the following command.

```bash

$ python vfn_train.py -h

```

## Evaluation

We provide the evaluation script to reproduce our evaluation results on [Flickr cropping dataset](https://github.com/yiling-chen/flickr-cropping-dataset). For example,

```bash

$ python vfn_eval.py --spp false --snapshot snapshots/model-wo-spp

```

You will need to get `sliding_window.json` and the test images from the [Flickr cropping dataset](https://github.com/yiling-chen/flickr-cropping-dataset) and specify the path of your model when running `vfn_eval.py`. You can also try our pre-trained model, which can be downloaded from [here](https://drive.google.com/drive/folders/0B0sDVRDPL5zBd3ozNlFmZEZpY1k?resourcekey=0-8LArgbYDQT07L-Dob6Yo9w&usp=sharing).

If you want to get an aesthetic score of a patch, please take a look at the example featured by [ModelDepot](https://modeldepot.io/yilingchen/view-finding-network)

## Questions?

If you have questions/suggestions, feel free to send an email to (yiling dot chen dot ntu at gmail dot com).

**If this work helps your research, please cite the following article:**

    @inproceedings{chen-acmmm-2017,

      title={Learning to Compose with Professional Photographs on the Web},

      author={Yi-Ling Chen and Jan Klopp and Min Sun and Shao-Yi Chien and Kwan-Liu Ma},

      booktitle={ACM Multimedia 2017},

      year={2017}

    }