A bag-of-regions approach to sketch-based image retrieval
A bag-of-regions approach to sketch-based image retrieval
复制标题
DOI：
10.1109/icip.2011.6116513
复制
发表时间：
2011-12
期刊：
影响因子：
--
通讯作者：
Rui Hu;T. Wang;J. Collomosse
中科院分区：
文献类型：
--
作者：
Rui Hu;T. Wang;J. Collomosse
本文提出了一个系统检索照片使用手绘草图查询。通过将分层图像分割的节点聚集到区域袋（BoR）表示中来从每个图像中提取区域。BoR表示多个尺度下的对象形状，即使在存在相邻杂波的情况下也对形状进行编码。我们从每个区域提取形状表示，使用梯度场HoG（GF-HOG）描述符，可以直接与草图查询进行比较。检索管道产生显着的性能改进，在以前的GF-HOG结果依赖于单尺度Canny边缘图，并超过领先的描述符（SIFT，SSIM）的视觉搜索。此外，我们的系统能够定位匹配图像内的素描对象。
This paper presents a system for retrieving photographs using free-hand sketched queries. Regions are extracted from each image by gathering nodes of a hierarchical image segmentation into a bag-of-regions (BoR) representation. The BoR represents object shape at multiple scales, encoding shape even in the presence of adjacent clutter. We extract a shape representation from each region, using the Gradient Field HoG (GF-HOG) descriptor which enables direct comparison with the sketched query. The retrieval pipeline yields significant performance improvements over the previous GF-HOG results reliant on single-scale Canny edge maps, and over leading descriptors (SIFT, SSIM) for visual search. In addition, our system enables localization of the sketched object within matching images.