Recent Advances in Zero-Shot Recognition: Toward Data-Efficient Understanding of Visual Content

With the recent renaissance of deep convolutional neural networks (CNNs), encouraging breakthroughs have been achieved on the supervised recognition tasks, where each class has sufficient and fully annotated training data. However, to scale the recognition to a large number of classes with few or no training samples for each class remains an unsolved problem. One approach is to develop models capable of recognizing unseen categories without any training instances, or zero-shot recognition/learning. This article provides a comprehensive review of existing zero-shot recognition techniques covering various aspects ranging from representations of models, data sets, and evaluation settings. We also overview related recognition tasks including one-shot and open-set recognition, which can be used as natural extensions of zero-shot recognition when a limited number of class samples become available or when zero-shot recognition is implemented in a real-world setting. We highlight the limitations of existing approaches and point out future research directions in this existing new research area.

• Publication year

  2017

• Venue

  IEEE Signal Processing Magazine

• Publication date

  2017-10-13

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1109/MSP.2017.2763441arXiv 1710.04837
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• From Zero-Shot Learning to Conventional Supervised Classification: Unseen Visual Data Synthesis

  2017cited by this paper
• Vocabulary-informed Extreme Value Learning

  2017cited by this paper
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks

  2017cited by this paper
• Zero-Shot Recognition Using Dual Visual-Semantic Mapping Paths

  2017cited by this paper
• Open Vocabulary Scene Parsing

  2017cited by this paper
• Zero-Shot Action Recognition with Error-Correcting Output Codes

  2017cited by this paper
• Zero-Shot Learning — The Good, the Bad and the Ugly

  2017influential reference
• Multi-task Deep Neural Network for Joint Face Recognition and Facial Attribute Prediction

  2017cited by this paper
• Video2vec Embeddings Recognize Events When Examples Are Scarce

  2017cited by this paper
• Joint Intermodal and Intramodal Label Transfers for Extremely Rare or Unseen Classes

  2017cited by this paper
• MOON: A Mixed Objective Optimization Network for the Recognition of Facial Attributes

  2016cited by this paper
• Walk and Learn: Facial Attribute Representation Learning from Egocentric Video and Contextual Data

  2016cited by this paper
• Progressive Neural Networks

  2016cited by this paper
• An Empirical Study and Analysis of Generalized Zero-Shot Learning for Object Recognition in the Wild

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

  2016cited by this paper
• Facial Attributes Classification Using Multi-task Representation Learning

  2016cited by this paper
• Dynamic Concept Composition for Zero-Example Event Detection

  2016cited by this paper
• iCaRL: Incremental Classifier and Representation Learning

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

  2016cited by this paper
• Learning to Learn: Model Regression Networks for Easy Small Sample Learning

  2016cited by this paper
• Understanding object descriptions in robotics by open-vocabulary object retrieval and detection

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

  2016cited by this paper
• Video Emotion Recognition with Transferred Deep Feature Encodings

  2016cited by this paper
• Fast Zero-Shot Image Tagging

  2016cited by this paper
• Synthesized Classifiers for Zero-Shot Learning

  2016influential reference
• Latent Embeddings for Zero-Shot Classification

  2016cited by this paper
• Concepts Not Alone: Exploring Pairwise Relationships for Zero-Shot Video Activity Recognition

  2016cited by this paper
• Learning feed-forward one-shot learners

  2016cited by this paper
• Learning from Small Sample Sets by Combining Unsupervised Meta-Training with CNNs

  2016cited by this paper
• Harnessing Object and Scene Semantics for Large-Scale Video Understanding

  2016cited by this paper
• One-shot Learning with Memory-Augmented Neural Networks

  2016cited by this paper
• Understanding Higher-Order Shape via 3D Shape Attributes

  2016cited by this paper
• Matching Networks for One Shot Learning

  2016cited by this paper
• End-to-End Localization and Ranking for Relative Attributes

  2016cited by this paper
• Semi-supervised Vocabulary-Informed Learning

  2016influential reference
• Towards Open-World Person Re-Identification by One-Shot Group-Based Verification

  2016cited by this paper
• Friction from Reflectance: Deep Reflectance Codes for Predicting Physical Surface Properties from One-Shot In-Field Reflectance

  2016cited by this paper
• Beyond Semantic Attributes: Discrete Latent Attributes Learning for Zero-Shot Recognition

  2016cited by this paper
• Recognizing an Action Using Its Name: A Knowledge-Based Approach

  2016cited by this paper
• Predicting Deep Zero-Shot Convolutional Neural Networks Using Textual Descriptions

  2015influential reference
• Zero-Shot Learning via Joint Latent Similarity Embedding

  2015influential reference
• Siamese Neural Networks for One-Shot Image Recognition

  2015cited by this paper
• Deep Convolutional Inverse Graphics Network

  2015cited by this paper
• Zero-shot object recognition by semantic manifold distance

  2015cited by this paper
• Objects2action: Classifying and Localizing Actions without Any Video Example

  2015cited by this paper
• Prediction of search targets from fixations in open-world settings

  2015cited by this paper
• Visual Domain Adaptation: A survey of recent advances

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

  2015cited by this paper
• Semi-Supervised Zero-Shot Classification with Label Representation Learning

  2015cited by this paper
• VideoStory Embeddings Recognize Events when Examples are Scarce

  2015cited by this paper
• Spatial Transformer Networks

  2015cited by this paper
• Lifelong Learning with Non-i.i.d. Tasks

  2015cited by this paper
• Semantic Concept Discovery for Large-Scale Zero-Shot Event Detection

  2015cited by this paper
• Transductive Zero-Shot Action Recognition by Word-Vector Embedding

  2015cited by this paper
• Ridge Regression, Hubness, and Zero-Shot Learning

  2015cited by this paper
• An embarrassingly simple approach to zero-shot learning

  2015cited by this paper
• Exploiting Feature and Class Relationships in Video Categorization with Regularized Deep Neural Networks

  2015cited by this paper
• Max-Margin Zero-Shot Learning for Multi-class Classification

  2015cited by this paper
• Transductive Multi-View Zero-Shot Learning

  2015influential reference
• ActivityNet: A large-scale video benchmark for human activity understanding

  2015cited by this paper
• Heterogeneous Knowledge Transfer in Video Emotion Recognition, Attribution and Summarization

  2015cited by this paper
• Human-level concept learning through probabilistic program induction

  2015cited by this paper
• Unsupervised Domain Adaptation for Zero-Shot Learning

  2015cited by this paper
• Hubness and Pollution: Delving into Cross-Space Mapping for Zero-Shot Learning

  2015cited by this paper
• COSTA: Co-Occurrence Statistics for Zero-Shot Classification

  2014cited by this paper
• Probability Models for Open Set Recognition

  2014cited by this paper
• Transductive Multi-label Zero-shot Learning

  2014influential reference
• Attributes Make Sense on Segmented Objects

  2014cited by this paper
• Going deeper with convolutions

  2014cited by this paper
• Improving zero-shot learning by mitigating the hubness problem

  2014cited by this paper
• GloVe: Global Vectors for Word Representation

  2014cited by this paper
• Zero-Shot Event Detection Using Multi-modal Fusion of Weakly Supervised Concepts

  2014cited by this paper
• Zero-shot recognition with unreliable attributes

  2014cited by this paper
• Transductive Multi-view Embedding for Zero-Shot Recognition and Annotation

  2014influential reference
• A Unified Perspective on Multi-Domain and Multi-Task Learning

  2014cited by this paper
• VideoStory: A New Multimedia Embedding for Few-Example Recognition and Translation of Events

  2014cited by this paper
• Towards Open World Recognition

  2014cited by this paper
• A Unified Semantic Embedding: Relating Taxonomies and Attributes

  2014cited by this paper
• Open-vocabulary Object Retrieval

  2014cited by this paper
• Return of the Devil in the Details: Delving Deep into Convolutional Nets

  2014cited by this paper
• Curriculum learning of multiple tasks

  2014cited by this paper
• Decorrelating Semantic Visual Attributes by Resisting the Urge to Share

  2014cited by this paper
• Inverse Graphics with Probabilistic CAD Models

  2014cited by this paper
• Evaluation of output embeddings for fine-grained image classification

  2014cited by this paper
• Large-Scale Object Classification Using Label Relation Graphs

  2014cited by this paper
• A PAC-Bayesian bound for Lifelong Learning

  2013cited by this paper
• Zero-Shot Learning Through Cross-Modal Transfer

  2013influential reference
• Distributed Representations of Words and Phrases and their Compositionality

  2013cited by this paper
• The Interestingness of Images

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

  2013influential reference
• NEIL: Extracting Visual Knowledge from Web Data

  2013cited by this paper
• Write a Classifier: Zero-Shot Learning Using Purely Textual Descriptions

  2013cited by this paper
• Label-Embedding for Attribute-Based Classification

  2013cited by this paper
• DeViSE: A Deep Visual-Semantic Embedding Model

  2013influential reference
• Transfer Learning in a Transductive Setting

  2013cited by this paper
• The Hubness Phenomenon: Fact or Artifact?

  2013cited by this paper
• Understanding and Predicting Interestingness of Videos

  2013cited by this paper
• Cumulative Attribute Space for Age and Crowd Density Estimation

  2013cited by this paper
• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition

  2013cited by this paper
• Efficient Estimation of Word Representations in Vector Space

  2013cited by this paper

• Concurrent historical data clustering and common feature learning for new-mode zero-shot industrial anomaly detection

  2026cites this paper
• GZSL-Lite: A Lightweight Generalized Zero-Shot Learning Network for SSVEP-Based BCIs

  2025cites this paper
• Cross-Modality Prompts: Few-Shot Multi-Label Recognition With Single-Label Training

  2025cites this paper
• Denoised and Dynamic Alignment Enhancement for Zero-Shot Learning

  2025cites this paper
• A knowledge transfer-based intelligent decision support method for fault management

  2025cites this paper
• Zero-shot gear fault diagnosis based on attribute consistency embedding

  2025cites this paper
• Weighted Mixture of Probabilistic PCA with Shared-Private Subspaces for Process Monitoring under New Operating Conditions

  2025cites this paper
• Segmentation-Based X-Ray Multiobjective Quality Assessment Network

  2025cites this paper
• Visual–Semantic Fuzzy Interaction Network for Zero-Shot Learning

  2025cites this paper
• Semantic Knowledge Base-Enabled Zero-Shot Multi-Level Feature Transmission Optimization

  2024cites this paper
• Zero-Shot Visual Sentiment Prediction via Cross-Domain Knowledge Distillation

  2024cites this paper
• GACRec: Generative adversarial contrastive learning for improved long-tail item recommendation

  2024cites this paper
• A Highly Efficient Zero-Shot Cross-Modal Hashing Method Based on CLIP

  2024cites this paper
• Rethinking attribute localization for zero-shot learning

  2024cites this paper
• Using large language models for extracting and pre-annotating texts on mental health from noisy data in a low-resource language

  2024cites this paper
• A survey and experimental study for embedding-aware generative models: Features, models, and any-shot scenarios

  2024cites this paper
• Feature Selection With Partial Autoencoding for Zero-Sample Fault Diagnosis

  2024cites this paper
• Zero shot plant disease classification with semantic attributes

  2024cites this paper
• Preserving Multi-Modal Capabilities of Pre-trained VLMs for Improving Vision-Linguistic Compositionality

  2024cites this paper
• Retrieval Instead of Fine-tuning: A Retrieval-based Parameter Ensemble for Zero-shot Learning

  2024cites this paper
• Beyond Adapter Retrieval: Latent Geometry-Preserving Composition via Sparse Task Projection

  2024cites this paper
• Anchor-based discriminative dual distribution calibration for transductive zero-shot learning

  2023cites this paper
• Robust Data Sampling in Machine Learning: A Game-Theoretic Framework for Training and Validation Data Selection

  2023cites this paper
• Class-Incremental Generalized Zero-Shot Learning

  2023cites this paper
• CorrEmbed: Evaluating Pre-trained Model Image Similarity Efficacy with a Novel Metric

  2023cites this paper
• Perspectives of Transfer Learning on the Diagnosis of Faults in Electrical Machines, Power Electronics, and Drives

  2023cites this paper
• From Image to Language: A Critical Analysis of Visual Question Answering (VQA) Approaches, Challenges, and Opportunities

  2023cites this paper
• How to Do Machine Learning with Small Data? - A Review from an Industrial Perspective

  2023cites this paper
• Machine Learning and Knowledge Graphs: Existing Gaps and Future Research Challenges

  2023cites this paper
• Episode-based training strategy for zero-shot semantic segmentation

  2023cites this paper
• Visual Content Analysis of STEM-Related Content in Indonesian Vocational High Schools Biology Textbooks

  2023cites this paper
• Relative attributes classification via transformers and rank SVM loss

  2023cites this paper
• JSSE: Joint Sequential Semantic Encoder for Zero-Shot Event Recognition

  2023cites this paper
• A Generalized Zero-Shot Learning Scheme for SSVEP-Based BCI System

  2023cites this paper
• Deep quantization network with visual-semantic alignment for zero-shot image retrieval

  2023cites this paper
• Intelligentization of Art Design System Based on Multidimensional Visual Image Reconstruction Algorithm

  2023cites this paper
• Semantic Prompt for Few-Shot Image Recognition

  2023cites this paper
• Indeterminate Probability Theory

  2023cites this paper
• A Systematic Evaluation and Benchmark for Embedding-Aware Generative Models: Features, Models, and Any-shot Scenarios

  2023cites this paper
• From Zero-Shot to Few-Shot Learning: A Step of Embedding-Aware Generative Models

  2023cites this paper
• Computer Vision and Image Understanding

  2022cites this paper
• A Novel MLCNN for Aircraft Target Classification in Optical Remote Sensing Images with Small Samples

  2022cites this paper
• Affective Embedding Framework with Semantic Representations from Tweets for Zero-Shot Visual Sentiment Prediction

  2022cites this paper
• Generative Adversarial Zero-Shot Learning for Cold-Start News Recommendation

  2022cites this paper
• Learning Attribute Features Based on Class Labels for SAR Ship Classification

  2022cites this paper
• Contributions of Shape, Texture, and Color in Visual Recognition

  2022cites this paper
• Polarity Loss: Improving Visual-Semantic Alignment for Zero-Shot Detection

  2022cites this paper
• Zero-Shot Audio Classification using Image Embeddings

  2022cites this paper
• A study on zero-shot learning from semantic viewpoint

  2022cites this paper
• A Generative Approach to Open Set Recognition Using Distance-Based Probabilistic Anomaly Augmentation

  2022cites this paper
• Meta-DZSL: a meta-dictionary learning based approach to zero-shot recognition

  2022influential citation
• Class-attribute inconsistency learning for novelty detection

  2022cites this paper
• Finger vein image inpainting using neighbor binary-wasserstein generative adversarial networks (NB-WGAN)

  2022cites this paper
• Low-resource Learning with Knowledge Graphs: A Comprehensive Survey

  2022cites this paper
• A Survey on Open Set Recognition

  2021cites this paper
• TransZero++: Cross Attribute-Guided Transformer for Zero-Shot Learning

  2021cites this paper
• Zero Shot on the Cold-Start Problem: Model-Agnostic Interest Learning for Recommender Systems

  2021cites this paper
• Recognition Awareness: An Application of Latent Cognizance to Open-Set Recognition

  2021cites this paper
• Norm-guided Adaptive Visual Embedding for Zero-Shot Sketch-Based Image Retrieval

  2021cites this paper
• JSE: Joint Semantic Encoder for zero-shot gesture learning

  2021cites this paper
• A Novel Framework Using Zero Shot Learning Technique for a Non-Factoid Question Answering System

  2021cites this paper
• AutoNovel: Automatically Discovering and Learning Novel Visual Categories

  2021cites this paper
• Conterfactual Generative Zero-Shot Semantic Segmentation

  2021cites this paper
• What Can Knowledge Bring to Machine Learning?—A Survey of Low-shot Learning for Structured Data

  2021cites this paper
• Zero-Shot Recognition based on Semantic Embeddings and Deep Clustering

  2021cites this paper
• Latent factor model based zero-shot learning

  2021cites this paper
• Reproducing the invention of a named reaction: zero-shot prediction of unseen chemical reactions.

  2021cites this paper
• An Efficient Framework for Zero-Shot Sketch-Based Image Retrieval

  2021cites this paper
• Multiscale Visual-Attribute Co-Attention for Zero-Shot Image Recognition

  2021cites this paper
• MODELESS: MODulation rEcognition with LimitEd SuperviSion

  2021cites this paper
• Hyperspectral Image Classification across Different Datasets: A Generalization to Unseen Categories

  2021cites this paper
• Low-resource Learning with Knowledge Graphs: A Comprehensive Survey

  2021cites this paper
• Zero-Shot Text Classification with Semantically Extended Graph Convolutional Network

  2021cites this paper
• Parallel Network to Learn Novelty from the Known

  2021cites this paper
• Zero shot augmentation learning in internet of biometric things for health signal processing

  2021cites this paper
• NucNormZSL: nuclear norm-based domain adaptation in zero-shot learning

  2021influential citation
• Incrementally Zero-Shot Detection by an Extreme Value Analyzer

  2021cites this paper
• Zero-Shot and Few-Shot Learning With Knowledge Graphs: A Comprehensive Survey

  2021cites this paper
• Evaluating Visual Properties via Robust HodgeRank

  2021cites this paper
• Prototypical Matching and Open Set Rejection for Zero-Shot Semantic Segmentation

  2021cites this paper
• Research on a small sample fault diagnosis method for a high-pressure common rail system

  2021cites this paper
• Recursive Training for Zero-Shot Semantic Segmentation

  2021cites this paper
• Knowledge-aware Zero-Shot Learning: Survey and Perspective

  2021cites this paper
• Class Knowledge Overlay to Visual Feature Learning for Zero-Shot Image Classification

  2021cites this paper
• Simple Multi-dataset Detection

  2021cites this paper
• Triple discriminator generative adversarial network for zero-shot image classification

  2021cites this paper
• Decoupling Zero-Shot Semantic Segmentation

  2021cites this paper
• Learning Object Attributes with Category-Free Grounded Language from Deep Featurization

  2020cites this paper
• Vocabulary-Informed Zero-Shot and Open-Set Learning

  2020cites this paper
• Towards zero-shot learning generalization via a cosine distance loss

  2020cites this paper
• Deep Learning for Free-Hand Sketch: A Survey

  2020cites this paper
• A Zero-Shot based Fingerprint Presentation Attack Detection System

  2020cites this paper
• Automatically Discovering and Learning New Visual Categories with Ranking Statistics

  2020cites this paper
• Active one-shot learning by a deep Q-network strategy

  2020cites this paper
• Class-Prototype Discriminative Network for Generalized Zero-Shot Learning

  2020cites this paper
• Model-Agnostic Metric for Zero-Shot Learning

  2020cites this paper
• Improved Visual-Semantic Alignment for Zero-Shot Object Detection

  2020cites this paper
• A Multi-Space Approach to Zero-Shot Object Detection

  2020cites this paper
• Few Shot Scene Classification in Remote Sensing using Meta-Agnostic Machine

  2020cites this paper
• Sifter: A Hybrid Workflow for Theme-based Video Curation at Scale

  2020cites this paper