Scalable Zero-Shot Learning via Binary Visual-Semantic Embeddings

Zero-shot learning aims to classify the visual instances from unseen classes in the absence of training examples. This is typically achieved by directly mapping visual features to a semantic embedding space of classes (e.g., attributes or word vectors), where the similarity between the two modalitie...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on image processing Jg. 28; H. 7; S. 3662 - 3674
Hauptverfasser: Shen, Fumin, Zhou, Xiang, Yu, Jun, Yang, Yang, Liu, Li, Shen, Heng Tao
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States IEEE 01.07.2019
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Algorithms
Bias
Binary codes
binary embeddings
Classification
Coding
Embedding
Exact solutions
Hamming distance
Image coding
Mapping
Mathematical analysis
Recognition
Semantics
Task analysis
Training
Visual discrimination
Visualization
Zero-shot learning
ISSN:1057-7149, 1941-0042, 1941-0042
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Zero-shot learning aims to classify the visual instances from unseen classes in the absence of training examples. This is typically achieved by directly mapping visual features to a semantic embedding space of classes (e.g., attributes or word vectors), where the similarity between the two modalities can be readily measured. However, the semantic space may not be reliable for recognition due to the noisy class embeddings or visual bias problem. In this paper, we propose a novel binary embedding-based zero-shot learning (BZSL) method, which recognizes the visual instances from unseen classes through an intermediate discriminative Hamming space. Specifically, BZSL jointly learns two binary coding functions to encode both visual instances and class embeddings into the Hamming space, which well alleviates the visual-semantic bias problem. As a desiring property, classifying an unseen instance thereby can be efficiently done by retrieving its nearest class codes with minimal Hamming distance. During training, by introducing two auxiliary variables for the coding functions, we formulate an equivalent correlation maximization problem, which admits an analytical solution. The resulting algorithm thus enjoys both highly efficient training and scalable novel class inferring. Extensive experiments on four benchmark datasets, including the full ImageNet Fall 2011 dataset with over 20k unseen classes, demonstrate the superiority of our method on the zero-shot learning task. Particularly, we show that increasing the binary embedding dimension can inevitably improve the recognition accuracy.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1057-7149
1941-0042
1941-0042
DOI:10.1109/TIP.2019.2899987