Abstract:Most of the existing algorithms for zero-shot classification problems typically rely on the attribute-based semantic relations among categories to realize the classification of novel categories without observing any of their instances. However, training the zero-shot classification models still requires attribute labeling for each class (or even instance) in the training dataset, which is also expensive. To this end, in this paper, we bring up a new problem scenario: "Can we derive zero-shot learning for novel attribute detectors/classifiers and use them to automatically annotate the dataset for labeling efficiency?". Basically, given only a small set of detectors that are learned to recognize some manually annotated attributes (i.e., the seen attributes), we aim to synthesize the detectors of novel attributes in a zero-shot learning manner. Our proposed method, Zero-Shot Learning for Attributes (ZSLA), which is the first of its kind to the best of our knowledge, tackles this new research problem by applying the set operations to first decompose the seen attributes into their basic attributes and then recombine these basic attributes into the novel ones. Extensive experiments are conducted to verify the capacity of our synthesized detectors for accurately capturing the semantics of the novel attributes and show their superior performance in terms of detection and localization compared to other baseline approaches. Moreover, we demonstrate the application of automatic annotation using our synthesized detectors on Caltech-UCSD Birds-200-2011 dataset. Various generalized zero-shot classification algorithms trained upon the dataset re-annotated by ZSLA show comparable performance with those trained with the manual ground-truth annotations. Please refer to our project page for source code: <a class="link-external link-https" href="https://yuhsuanli.github.io/ZSLA/" rel="external noopener nofollow">this https URL</a>

Zero-Shot Learning with Attribute Selection.

Selective Zero-Shot Classification with Augmented Attributes.

Learning Discriminative Latent Attributes for Zero-Shot Classification.

Improving Generalization via Attribute Selection on Out-of-the-box Data

Zero-shot Learning with a Partial Set of Observed Attributes

Zero-Shot Learning via Attribute Regression and Class Prototype Rectification

Prior Knowledge about Attributes: Learning a More Effective Potential Space for Zero-Shot Recognition

Visual-guided attentive attributes embedding for zero-shot learning

Estimation of Near-Instance-Level Attribute Bottleneck for Zero-Shot Learning

Attribute self-representation steered by exclusive lasso for zero-shot learning

Boosting Zero-shot Learning via Contrastive Optimization of Attribute Representations

Attribute subspaces for zero-shot learning

OntoZSL: Ontology-enhanced Zero-shot Learning

Make an Omelette with Breaking Eggs: Zero-Shot Learning for Novel Attribute Synthesis

TransZero++: Cross Attribute-Guided Transformer for Zero-Shot Learning

Landmark Selection for Zero-shot Learning.

ZS-VAT: Learning Unbiased Attribute Knowledge for Zero-Shot Recognition Through Visual Attribute Transformer

Complementary Attributes: A New Clue to Zero-Shot Learning

A Large-Scale Attribute Dataset for Zero-Shot Learning

TransZero: Attribute-guided Transformer for Zero-Shot Learning

Learn More from Less: Generalized Zero-Shot Learning with Severely Limited Labeled Data