SBIR regression

This repo contains code for the C&G paper "Sketching out the details: Sketch-based image retrieval using convolutional neural networks with multi-stage regression"

Dependencies

You will need to compile Caffe v1.0 with customized L2 normalize layer. Check caffe_utils/README.md for instructions.

Alternatively, you can use standard Caffe, just remove the normalize layer in model/*.prototxt, then normalise the output manually using e.g. numpy.

Pretrained model

Pretrained model (and dataset) can be downloaded here.

Feature extraction

Check getfeat_img.py and getfeat_skt.py for examples of extracting features from a raw image/sketch.

Demo

Reference

@article{bui2018sketching,
  title={Sketching out the Details: Sketch-based Image Retrieval using Convolutional Neural Networks with Multi-stage Regression},
  author={Bui, Tu and Ribeiro, Leonardo and Ponti, Moacir and Collomosse, John},
  journal={Computers \& Graphics},
  year={2018},
  publisher={Elsevier}
}