arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1804.04412cs.CV
收藏

作为结构表示的对象地标的无监督发现

Unsupervised Discovery of Object Landmarks as Structural Representations

University of Michigan, Ann Arbor(密歇根大学安娜堡分校);Google Brain(谷歌大脑)

Yuting Zhang, Yijie Guo, Yixin Jin, Yijun Luo, Zhiyuan He, Honglak Lee

摘要:

深度神经网络可以用丰富的潜在表征对图像进行建模,但它们不能以人类可感知的方式自然地概念化对象类别的结构。本文讨论了在没有监督的图像建模过程中学习对象结构的问题。我们提出了一个自动编码公式来发现地标作为显式的结构表示。编码模块输出地标坐标,其有效性由反映地标的必要属性的约束来确保。解码模块将地标作为端到端可微分框架中的可学习输入表示的一部分。我们发现的地标是语义上有意义的,更预测手动注释的地标比以前的方法发现的。地标的坐标也是预训练的深度神经网络表示在识别视觉属性时的补充特征。此外,所提出的方法自然地创建了一个无监督的,可感知的界面,以操纵对象的形状和解码具有可控结构的图像。该项目的网页是http://ytzhang.net/projects/lmdis-rep

英文摘要:

Deep neural networks can model images with rich latent representations, but they cannot naturally conceptualize structures of object categories in a human-perceptible way. This paper addresses the problem of learning object structures in an image modeling process without supervision. We propose an autoencoding formulation to discover landmarks as explicit structural representations. The encoding module outputs landmark coordinates, whose validity is ensured by constraints that reflect the necessary properties for landmarks. The decoding module takes the landmarks as a part of the learnable input representations in an end-to-end differentiable framework. Our discovered landmarks are semantically meaningful and more predictive of manually annotated landmarks than those discovered by previous methods. The coordinates of our landmarks are also complementary features to pretrained deep-neural-network representations in recognizing visual attributes. In addition, the proposed method naturally creates an unsupervised, perceptible interface to manipulate object shapes and decode images with controllable structures. The project webpage is at http://ytzhang.net/projects/lmdis-rep

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。

↑