We propose and evaluate several triplet CNN architectures for measuring the
similarity between sketches and photographs, within the context of the sketch
based image retrieval (SBIR) task. In contrast to recent fine-grained SBIR
work, we study the ability of our networks to generalise across diverse object
categories from limited training data, and explore in detail strategies for
weight sharing, pre-processing, data augmentation and dimensionality reduction.
We exceed the performance of pre-existing techniques on both the Flickr15k
category level SBIR benchmark by $18\%$, and the TU-Berlin SBIR benchmark by
$\sim10 \mathcal{T}_b$, when trained on the 250 category TU-Berlin
classification dataset augmented with 25k corresponding photographs harvested
from the Internet.

本研究提出和评估了几种三元 CNN 结构，用于在素描和照片之间测量相似度，重点研究网络在有限的训练数据下的分类能力，探究了权值共享、预处理、数据增强和降维等策略，并在 250 个 TU-Berlin 分类数据集上进行训练，结合从互联网上采集的 25,000 张对应照片，相较于现有技术在 Flickr15k 和 TU-Berlin SBIR 基准上分别提升了 18% 和 10 个标准差。