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.
Temporal Context Aggregation Network for Temporal Action Proposal Refinement
The Blessings of Unlabeled Background in Untrimmed Videos
Learning without Seeing nor Knowing: Towards Open Zero-Shot Learning
Incrementally Zero-Shot Detection by an Extreme Value Analyzer
Multimodal Motion Prediction with Stacked Transformers
AdaSGN: Adapting Joint Number and Model Size for Efficient Skeleton-Based Action Recognition
Cyber-Attack Detection in Socio-Technical Transportation Systems Exploiting Redundancies Between Physical and Social Data
MoViNets: Mobile Video Networks for Efficient Video Recognition
Classifier Crafting: Turn Your ConvNet into a Zero-Shot Learner!
PSCC-Net: Progressive Spatio-Channel Correlation Network for Image Manipulation Detection and Localization
Learning Multiscale Correlations for Human Motion Prediction
Online Lifelong Generalized Zero-Shot Learning
DFS: A Diverse Feature Synthesis Model for Generalized Zero-Shot Learning
Large-Scale Zero-Shot Image Classification from Rich and Diverse Textual Descriptions
Aggregated Multi-GANs for Controlled 3D Human Motion Prediction
Skeleton Aware Multi-modal Sign Language Recognition
Decision Making of Connected Automated Vehicles at An Unsignalized Roundabout Considering Personalized Driving Behaviours
Multilingual Code-Switching for Zero-Shot Cross-Lingual Intent Prediction and Slot Filling
Cooperative Self-training of Machine Reading Comprehension
Self-Supervised Motion Retargeting with Safety Guarantee
Temporal Action Segmentation from Timestamp Supervision
ACTION-Net: Multipath Excitation for Action Recognition
PcmNet: Position-Sensitive Context Modeling Network for Temporal Action Localization
Understanding the Robustness of Skeleton-based Action Recognition under Adversarial Attack
On complementing end-to-end human behavior predictors with planning
On Implicit Attribute Localization for Generalized Zero-Shot Learning
Simple online and real-time tracking with occlusion handling
Goal-Oriented Gaze Estimation for Zero-Shot Learning
FloMo: Tractable Motion Prediction with Normalizing Flows
Effects of Image Compression on Face Image Manipulation Detection: A Case Study on Facial Retouching
Modeling Multi-Label Action Dependencies for Temporal Action Localization
Task Aligned Generative Meta-learning for Zero-shot Learning
Domain and View-point Agnostic Hand Action Recognition
Exploiting latent representation of sparse semantic layers for improved short-term motion prediction with Capsule Networks
Counterfactual Zero-Shot and Open-Set Visual Recognition
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.