LOCAL-GLOBAL REPRESENTATION LEARNING FOR ON-THE-FLY FG-SBIR
demo.mp4
This repository is the official pytorch implementation of our paper, *LOCAL-GLOBAL REPRESENTATION LEARNING FOR ON-THE-FLY FG-SBIR*.
We use two standard datasets as follows:
The QMUL-Shoe-V2 dataset contains 6730 sketches and 2000 photos. A total of 6051 sketches and 1800 photos were used to train the models, and the rest were used to test the models.
The QMUL-Chair-V2 dataset contains 2000 sketches and 400 photos. A total of 1275 sketches and 300 photos were used to train the models, and the rest were used to test the models.
-
Release training code
-
Release testing code
-
Release pre-trained models
coming soon......
We would like to thank all of reviewers for their constructive comments and CQUPT for supporting.
This repo is currently maintained by Dawei Dai (dw_dai@163.com) and his master's student Yingge Liu (s200231105@stu.cqupt.edu.cn).