Bi-Directional Distribution Alignment for Transductive Zero-Shot Learning

Zero-shot learning (ZSL) suffers intensely from the domain shift issue, i.e., the mismatch (or misalignment) between the true and learned data distributions for classes without training data (unseen classes). By learning additionally from unlabelled data collected for the unseen classes, transductive ZSL (TZSL) could reduce the shift but only to a certain extent. To improve TZSL, we propose a novel approach Bi-VAEGAN which strengthens the distribution alignment between the visual space and an auxiliary space. As a result, it can reduce largely the domain shift. The proposed key designs include (1) a bi-directional distribution alignment, (2) a simple but effective L2-norm based feature normalization approach, and (3) a more sophisticated unseen class prior estimation. Evaluated by four benchmark datasets, Bi-VAEGAN11Code is available at https://github.com/Zhicaiwww/Bi-VAEGAN achieves the new state of the art under both the standard and generalized TZSL settings.

• Publication year

  2023

• Venue

  Computer Vision and Pattern Recognition

• Publication date

  2023-03-15

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1109/CVPR52729.2023.01905arXiv 2303.08698
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Iterative Class Prototype Calibration for Transductive Zero-Shot Learning

  2023cited by this paper
• MSDN: Mutually Semantic Distillation Network for Zero-Shot Learning

  2022cited by this paper
• En-Compactness: Self-Distillation Embedding & Contrastive Generation for Generalized Zero-Shot Learning

  2022cited by this paper
• Siamese Contrastive Embedding Network for Compositional Zero-Shot Learning

  2022cited by this paper
• Learning semantic ambiguities for zero-shot learning

  2022cited by this paper
• Non-generative Generalized Zero-shot Learning via Task-correlated Disentanglement and Controllable Samples Synthesis

  2022cited by this paper
• Visual Structure Constraint for Transductive Zero-Shot Learning in the Wild

  2021cited by this paper
• OntoZSL: Ontology-enhanced Zero-shot Learning

  2021cited by this paper
• Adaptive and Generative Zero-Shot Learning

  2021cited by this paper
• Hardness Sampling for Self-Training Based Transductive Zero-Shot Learning

  2021influential reference
• Adversarial Feature Hallucination Networks for Few-Shot Learning

  2020cited by this paper
• On Feature Normalization and Data Augmentation

  2020cited by this paper
• A Review of Generalized Zero-Shot Learning Methods

  2020cited by this paper
• Self-Supervised Domain-Aware Generative Network for Generalized Zero-Shot Learning

  2020cited by this paper
• Domain Adaptation with Conditional Distribution Matching and Generalized Label Shift

  2020cited by this paper
• Latent Embedding Feedback and Discriminative Features for Zero-Shot Classification

  2020influential reference
• A Unified View of Label Shift Estimation

  2020cited by this paper
• Semantically Aligned Bias Reducing Zero Shot Learning

  2019cited by this paper
• Gradient Matching Generative Networks for Zero-Shot Learning

  2019cited by this paper
• Rethinking Zero-Shot Learning: A Conditional Visual Classification Perspective

  2019cited by this paper
• Transductive Zero-Shot Learning with Visual Structure Constraint

  2019cited by this paper
• F-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learning

  2019influential reference
• Creativity Inspired Zero-Shot Learning

  2019cited by this paper
• Leveraging the Invariant Side of Generative Zero-Shot Learning

  2019cited by this paper
• Stacked Semantics-Guided Attention Model for Fine-Grained Zero-Shot Learning

  2018cited by this paper
• Detecting and Correcting for Label Shift with Black Box Predictors

  2018cited by this paper
• Multi-modal Cycle-consistent Generalized Zero-Shot Learning

  2018cited by this paper
• Progressive Ensemble Networks for Zero-Shot Recognition

  2018cited by this paper
• Zero-Shot Classification with Discriminative Semantic Representation Learning

  2017cited by this paper
• Wasserstein Generative Adversarial Networks

  2017influential reference
• Zero-Shot Visual Recognition Using Semantics-Preserving Adversarial Embedding Networks

  2017cited by this paper
• Learning to Compare: Relation Network for Few-Shot Learning

  2017cited by this paper
• Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly

  2017influential reference
• Multi-label Zero-Shot Learning with Structured Knowledge Graphs

  2017cited by this paper
• Improved Training of Wasserstein GANs

  2017cited by this paper
• A Simple Exponential Family Framework for Zero-Shot Learning

  2017cited by this paper
• Generative Adversarial Networks: An Overview

  2017cited by this paper
• Feature Generating Networks for Zero-Shot Learning

  2017influential reference
• Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks

  2017cited by this paper
• Tutorial on Variational Autoencoders

  2016cited by this paper
• Learning Deep Representations of Fine-Grained Visual Descriptions

  2016cited by this paper
• Learning a Deep Embedding Model for Zero-Shot Learning

  2016cited by this paper
• Zero-Shot Recognition via Structured Prediction

  2016cited by this paper
• Predicting Visual Exemplars of Unseen Classes for Zero-Shot Learning

  2016cited by this paper
• Transductive Multi-View Zero-Shot Learning

  2015cited by this paper
• Deep Residual Learning for Image Recognition

  2015influential reference
• Label-Embedding for Attribute-Based Classification

  2013cited by this paper
• Zero-Shot Learning by Convex Combination of Semantic Embeddings

  2013cited by this paper
• SUN attribute database: Discovering, annotating, and recognizing scene attributes

  2012influential reference
• The Caltech-UCSD Birds-200-2011 Dataset

  2011cited by this paper
• Caltech-UCSD Birds 200

  2010cited by this paper
• Zero-data Learning of New Tasks

  2008cited by this paper
• Attribute-Based Classification for Zero-Shot Visual Object Categorization

  year unknowncited by this paper

• AugGen: a generative framework for continual generalized zero-shot learning

  2026cites this paper
• ZeroDiff++: Substantial Unseen Visual-semantic Correlation in Zero-shot Learning

  2026cites this paper
• Hybrid Granularity Distribution Estimation for Few-Shot Learning: Statistics Transfer From Categories and Instances

  2026cites this paper
• Dual Attention-Guided GAN for transductive Zero-Shot fault diagnosis of nuclear power seal

  2026cites this paper
• Generative Model-Based Mixed-Semantic Enhancement for Transductive Zero-Shot Learning

  2026cites this paper
• A Multi-Modal Knowledge-Driven Approach for Generalized Zero-shot Video Classification

  2026cites this paper
• Self-Assembled Generative Framework for Generalized Zero-Shot Learning

  2025cites this paper
• CoCNet: A Chain-of-Clues framework for zero-shot referring expression comprehension

  2025cites this paper
• Progressive Feature-Attribute Matching via Bi-Directional Generation for Transductive Zero-Shot Learning

  2025influential citation
• A shapelet-driven distillation generation method for generalized zero-shot learning in compound fault diagnosis

  2025cites this paper
• Multi-Granularity Mutual Refinement Network for Zero-Shot Learning

  2025cites this paper
• Leveraging Self-Distillation and Disentanglement Network to Enhance Visual–Semantic Feature Consistency in Generalized Zero-Shot Learning

  2024cites this paper
• Cross-modal zero-sample diagnosis framework utilizing non-contact sensing data fusion

  2024cites this paper
• Enhance Image Classification via Inter-Class Image Mixup with Diffusion Model

  2024cites this paper
• Estimation of Near-Instance-Level Attribute Bottleneck for Zero-Shot Learning

  2024cites this paper
• ZeroDiff: Solidified Visual-semantic Correlation in Zero-Shot Learning

  2024cites this paper
• M-RRFS: A Memory-Based Robust Region Feature Synthesizer for Zero-Shot Object Detection

  2024influential citation
• Zero-Shot Image Classification: Recent Status and Future Trends

  2024cites this paper
• Enhancing Zero-Shot Vision Models by Label-Free Prompt Distribution Learning and Bias Correcting

  2024cites this paper
• GLCM-Adapter: Global-Local Content Matching for Few-shot CLIP Adaptation

  2024cites this paper
• Locate and Verify: A Two-Stream Network for Improved Deepfake Detection

  2023cites this paper