Attentive Region Embedding Network for Zero-Shot Learning

By Guosen Xie, Li Liu, Xiaobo Jin, Fan Zhu, Zheng Zhang, Jie Qin, Yazhou Yao, Ling Shao. 2019

297 citations recorded by the source corpus.

Zero-shot learning (ZSL) aims to classify images from unseen categories, by merely utilizing seen class images as the training data. Existing works on ZSL mainly leverage the global features or learn the global regions, from which, to construct the embeddings to the semantic space. However, few of them study the discrimination power implied in local image regions (parts), which, in some sense, correspond to semantic attributes, have stronger discrimination than attributes, and can thus assist the semantic transfer between seen/unseen classes. In this paper, to discover (semantic) regions, we propose the attentive region embedding network (AREN), which is tailored to advance the ZSL task. Specifically, AREN is end-to-end trainable and consists of two network branches, i.e., the attentive region embedding (ARE) stream, and the attentive compressed second-order embedding (ACSE) stream. ARE is capable of discovering multiple part regions under the guidance of the attention and the compatibility loss. Moreover, a novel adaptive thresholding mechanism is proposed for suppressing redundant (such as background) attention regions. To further guarantee more stable semantic transfer from the perspective of second-order collaboration, ACSE is incorporated into the AREN. In the comprehensive evaluations on four benchmarks, our models achieve state-of-the-art performances under ZSL setting, and compelling results under generalized ZSL setting.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Detection and Continual Learning of Novel Face Presentation Attacks (2021, background): “The learning setting that we explore is more related to the zero-shot learning (ZSL) formulation [9, 37, 38, 25].”
Attribute Prototype Network for Zero-Shot Learning (2020, background): “, SGMA [58], AREN [46], LFGAA+Hybrid [27]; and generative models i.”
Video Object Segmentation with Episodic Graph Memory Networks (2020, background): “mory networks augment neural networks with an external memory component[28,43,58], which allow the network to explicitly access the past experiences. They have been shown eective in few-shot learning[39,62,63] and object tracking [67]. Recently, episodic external memory networks have been explored t”
Region Graph Embedding Network for Zero-Shot Learning (2020, methodology): “Moreover, to address the domain bias issue (Fig.1(a)-(b)) in the end-to-end GZSL models [52, 80, 79], we propose a balance loss by minimizing the maximum response consistency between seen and unseen predictions.”
Contrastive Embedding for Generalized Zero-Shot Learning (2021, methodology): “To relieve the overfitting problem, some methods [10, 44, 3, 29, 73, 74, 49] have designed new loss functions to balance the predictions between seen and unseen classes.”
Adaptive and Generative Zero-Shot Learning (2021, methodology): “After adding IAS to our model, both seen and unseen accuracy increases, and the harmonic mean H improves about 1.5 % on average.”