Dating Historical Yearbook Photos

This is the reference implementation of the dating experiments from the updated version of https://arxiv.org/abs/1511.02575:

A Century of Portraits: A Visual Historical Record of American High School Yearbooks
Shiry Ginosar, Kate Rakelly, Sarah M. Sachs, Brian Yin, Crystal Lee, Philipp Krähenbühl and Alexei A. Efros
arXiv: Coming soon!

Contents

• src: net, solver, and data layer specifications to train Yearbook dating models
  • yearbook_layers.py: a Python data layer that loads images and feeds them into a Caffe network
  • solver.py: set solver parameters in Python
  • net.py: NetSpec defines the network architecture
  • solve.py: training script
• notebooks: visualize training curves, evaluate trained networks and reproduce paper figures
• data: data splits
• caffe: the Caffe framework, included as a git submodule pointing to a compatible version

License

This project is licensed for open non-commercial distribution under the UC Regents license; see LICENSE. Its dependencies, such as Caffe, are subject to their own respective licenses.

Requirements & Installation

Caffe, Python, and Jupyter are necessary for all of the experiments. Any installation or general Caffe inquiries should be directed to the caffe-users mailing list.

1. After cloning this repository, do git submodule init and git submodule update inside the Caffe directory to clone the caffe submodule.
2. Install Caffe dependencies. See the installation guide and try Caffe through Docker (recommended). Make sure to configure pycaffe, the Caffe Python interface, too.
3. Follow Caffe installation instructions to install required Python packages. Miniconda comes with many of the requirements (recommended).
4. Install Jupyter, the interface for viewing, executing, and altering the notebooks.
5. Configure your PYTHONPATH as indicated by the included .envrc so that this project dir and pycaffe are included.
6. Download the Yearbook dataset from the project page. Place the women dataset in data/faces/women/images and the men dataset in data/faces/men/images.

If you don't want to train your own model, the Python notebooks can be used for inference with our trained dating models To train a dating model, download the model weights for VGG-16 pre-trained on ILSVRC and place them in a directory called models. Configure training settings in train_example.sh, then call this script to train a dating network.