arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1908.06537cs.CV
收藏

超像素流:基于多层神经特征的语义对应

Hyperpixel Flow: Semantic Correspondence with Multi-layer Neural Features

POSTECH(浦项科技大学);Neural Processing Research Center(神经处理研究中心);Inria(法国国家信息与自动化研究所);ENS(巴黎高等师范学院);CNRS(法国国家科学研究中心);PSL University(巴黎文理研究大学)

Juhong Min, Jongmin Lee, Jean Ponce, Minsu Cho

摘要:

在大的类内变化下建立视觉对应关系需要在不同层次上分析图像,从与语义和上下文相关的特征到局部模式,同时对特定于实例的细节保持不变。为了应对这些挑战,我们用“超像素”来表示图像,这些超像素利用了从卷积神经网络的早期到晚期层中选择的少量相关特征。利用超像素的压缩特性,提出了一种基于Hough几何投票的实时匹配算法。所提出的方法,超像素流,在三个标准基准以及一个新的数据集SPair-71 k上设置了一个新的技术状态,该数据集包含比现有数据集更多的图像对,具有更准确和更丰富的注释以进行深入分析。

英文摘要:

Establishing visual correspondences under large intra-class variations requires analyzing images at different levels, from features linked to semantics and context to local patterns, while being invariant to instance-specific details. To tackle these challenges, we represent images by "hyperpixels" that leverage a small number of relevant features selected among early to late layers of a convolutional neural network. Taking advantage of the condensed features of hyperpixels, we develop an effective real-time matching algorithm based on Hough geometric voting. The proposed method, hyperpixel flow, sets a new state of the art on three standard benchmarks as well as a new dataset, SPair-71k, which contains a significantly larger number of image pairs than existing datasets, with more accurate and richer annotations for in-depth analysis.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑