Zero-Shot Learning
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 7
- 평가표
- 33
- 보존된 평가 기록
- 84
Zero-shot learning (ZSL) is a model's ability to detect classes never seen during training. The condition is that the classes are not known during supervised learning.
Earlier work in zero-shot learning use attributes in a two-step approach to infer unknown classes. In the computer vision context, more recent advances learn mappings from image feature space to semantic space. Other approaches learn non-linear multimodal embeddings. In the modern NLP context, language models can be evaluated on downstream tasks without fine tuning.
Benchmark datasets for zero-shot learning include aPY, AwA, and CUB, among others.
( Image credit: Prototypical Networks for Few shot Learning in PyTorch )
Further readings:
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
Attribute Learning for Network Intrusion Detection
Improving Semantic Embedding Consistency by Metric Learning for Zero-Shot Classification
Emotion-Based Crowd Representation for Abnormality Detection
Dynamic Probabilistic Network Based Human Action Recognition
Spatio-Temporal LSTM with Trust Gates for 3D Human Action Recognition
Kinematic-Layout-aware Random Forests for Depth-based Action Recognition
Multi-Camera Action Dataset for Cross-Camera Action Recognition Benchmarking
Hierarchical Attention Network for Action Recognition in Videos
Transition Forests: Learning Discriminative Temporal Transitions for Action Recognition and Detection
Annotation Methodologies for Vision and Language Dataset Creation
Action Recognition with Joint Attention on Multi-Level Deep Features
Collaborative Training of Tensors for Compositional Distributional Semantics
Zero-Shot Visual Recognition via Bidirectional Latent Embedding
VideoLSTM Convolves, Attends and Flows for Action Recognition
Zero-Shot Learning with Multi-Battery Factor Analysis
Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering
A Hierarchical Pose-Based Approach to Complex Action Understanding Using Dictionaries of Actionlets and Motion Poselets
Efficient Pairwise Learning Using Kernel Ridge Regression: an Exact Two-Step Method
Hand Action Detection from Ego-centric Depth Sequences with Error-correcting Hough Transform
Multi-Label Zero-Shot Learning via Concept Embedding
Fast Zero-Shot Image Tagging
Semi-supervised Zero-Shot Learning by a Clustering-based Approach
Learning Latent Sub-events in Activity Videos Using Temporal Attention Filters
Predicting Visual Exemplars of Unseen Classes for Zero-Shot Learning
Neighborhood Sensitive Mapping for Zero-Shot Classification using Independently Learned Semantic Embeddings
Action Classification via Concepts and Attributes
Learning Deep Representations of Fine-grained Visual Descriptions
Going Deeper into Action Recognition: A Survey
An Empirical Study and Analysis of Generalized Zero-Shot Learning for Object Recognition in the Wild
Going Deeper into First-Person Activity Recognition
Action Recognition in Video Using Sparse Coding and Relative Features
Revisiting Human Action Recognition: Personalization vs. Generalization
Dominant Codewords Selection with Topic Model for Action Recognition
Improved Dense Trajectory with Cross Streams
Real-time Action Recognition with Enhanced Motion Vector CNNs
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.