.png){kind=icon}
.png){kind=icon}
Return
Hierarchical disentangling network for object representation learning
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2023.109539.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
An object can be described as the combination of primary visual attributes. Disentangling such under-lying primitives is the long-term objective of representation learning. It is observed that categories have natural hierarchical characteristics, i.e., any two objects can share some common primitives at a partic-ular category level while possess unique traits at another. However, previous works usually operate in a flat manner (i.e., at a particular level) to disentangle the representations of objects. Even though they may obtain the primitives to constitute objects as the categories at that level, their results are obvi-ously not efficient and complete. In this paper, we propose a Hierarchical Disentangling Network (HDN) to exploit the rich hierarchical characteristics among categories to divide the disentangling process in a coarse-to-fine manner (i.e., level-wise), such that each level only focuses on learning the specific rep-resentations and finally the common and unique representations at all levels jointly constitute the raw object. Specifically, HDN is designed based on an encoder-decoder architecture. To simultaneously ensure the level-wise disentanglement and interpretability of the encoded representations, a novel hierarchical Generative Adversarial Network (GAN) is introduced. Quantitative and qualitative evaluations on popular object datasets validate the effectiveness of our method. (c) 2023 Published by Elsevier Ltd.
Keywords:
Object understanding
Hierarchical learning
Representation disentanglement
Generative adversarial network
Network interpretability
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W