Transductive Zero-Shot Learning with Visual Structure Constraint

Abstract

To recognize objects of the unseen classes, most existing Zero-Shot Learning(ZSL) methods first learn a compatible projection function between the common semantic space and the visual space based on the data of source seen classes, then directly apply it to the target unseen classes. However, in real scenarios, the data distribution between the source and target domain might not match well, thus causing the well-known \textbf{domain shift} problem. Based on the observation that visual features of test instances can be separated into different clusters, we propose a new visual structure constraint on class centers for transductive ZSL, to improve the generality of the projection function (i.e. alleviate the above domain shift problem). Specifically, three different strategies (symmetric Chamfer-distance, Bipartite matching distance, and Wasserstein distance) are adopted to align the projected unseen semantic centers and visual cluster centers of test instances. We also propose a new training strategy to handle the real cases where many unrelated images exist in the test dataset, which is not considered in previous methods. Experiments on many widely used datasets demonstrate that the proposed visual structure constraint can bring substantial performance gain consistently and achieve state-of-the-art results. The source code is available at \url{https://github.com/raywzy/VSC}.

NeurIPS 2019, code available at https://github.com/raywzy/VSC

References in corpus (11)

• Label-Embedding for Image Classification
• Zero-Shot Learning by Convex Combination of Semantic Embeddings
• Transductive Multi-view Zero-Shot Learning
• Learning Deep Representations of Fine-grained Visual Descriptions
• Semantic-Guided Multi-Attention Localization for Zero-Shot Learning
• Preserving Semantic Relations for Zero-Shot Learning
• Synthesized Classifiers for Zero-Shot Learning
• Zero-Shot Learning via Class-Conditioned Deep Generative Models
• Unsupervised Learning on Neural Network Outputs: with Application in Zero-shot Learning
• Discriminative Learning of Latent Features for Zero-Shot Recognition
• Transductive Unbiased Embedding for Zero-Shot Learning

Cited by in corpus (3)

• Joint one-sided synthetic unpaired image translation and segmentation for colorectal cancer prevention
• Transfer feature generating networks with semantic classes structure for zero-shot learning
• KMF: Knowledge-Aware Multi-Faceted Representation Learning for Zero-Shot Node Classification