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.
TAFE-Net: Task-Aware Feature Embeddings for Low Shot Learning
Advanced Customer Activity Prediction based on Deep Hierarchic Encoder-Decoders
On zero-shot recognition of generic objects
Localized Trajectories for 2D and 3D Action Recognition
Attentive Action and Context Factorization
Action Recognition from Single Timestamp Supervision in Untrimmed Videos
Knowledge Distillation for Human Action Anticipation
Leveraging the Invariant Side of Generative Zero-Shot Learning
SCSampler: Sampling Salient Clips from Video for Efficient Action Recognition
What and How Well You Performed? A Multitask Learning Approach to Action Quality Assessment
Context-aware Human Motion Prediction
LP-3DCNN: Unveiling Local Phase in 3D Convolutional Neural Networks
Semantic-Aware Knowledge Preservation for Zero-Shot Sketch-Based Image Retrieval
SMURFF: a High-Performance Framework for Matrix Factorization
End-to-End Video Captioning
75 Languages, 1 Model: Parsing Universal Dependencies Universally
Semantics-Guided Neural Networks for Efficient Skeleton-Based Human Action Recognition
Multigrid Predictive Filter Flow for Unsupervised Learning on Videos
Creativity Inspired Zero-Shot Learning
Adaptive Adjustment with Semantic Feature Space for Zero-Shot Recognition
RefineLoc: Iterative Refinement for Weakly-Supervised Action Localization
Skip prediction using boosting trees based on acoustic features of tracks in sessions
Zero-shot Image Recognition Using Relational Matching, Adaptation and Calibration
INFER: INtermediate representations for FuturE pRediction
f-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learning
Unifying Unsupervised Domain Adaptation and Zero-Shot Visual Recognition
On the Importance of Video Action Recognition for Visual Lipreading
Modelling Sequential Music Track Skips using a Multi-RNN Approach
What the Constant Velocity Model Can Teach Us About Pedestrian Motion Prediction
Cross-task weakly supervised learning from instructional videos
Spatiotemporal Filtering for Event-Based Action Recognition
MFAS: Multimodal Fusion Architecture Search
Zero Shot Learning with the Isoperimetric Loss
Two-Stream Action Recognition-Oriented Video Super-Resolution
Tracking without bells and whistles
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.