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.
Semi-supervised Vocabulary-informed Learning
Convolutional Two-Stream Network Fusion for Video Action Recognition
Kernelized Covariance for Action Recognition
The THUMOS Challenge on Action Recognition for Videos "in the Wild"
Improving Human Action Recognition by Non-action Classification
Online Human Action Detection using Joint Classification-Regression Recurrent Neural Networks
Long-term Temporal Convolutions for Action Recognition
NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis
Trajectory Aligned Features For First Person Action Recognition
Hollywood in Homes: Crowdsourcing Data Collection for Activity Understanding
Less is more: zero-shot learning from online textual documents with noise suppression
Learning a Pose Lexicon for Semantic Action Recognition
Tensor Representations via Kernel Linearization for Action Recognition from 3D Skeletons (Extended Version)
Multi-Cue Zero-Shot Learning with Strong Supervision
Latent Embeddings for Zero-shot Classification
Shuffle and Learn: Unsupervised Learning using Temporal Order Verification
Co-occurrence Feature Learning for Skeleton based Action Recognition using Regularized Deep LSTM Networks
Deep Multimodal Feature Analysis for Action Recognition in RGB+D Videos
Multi-velocity neural networks for gesture recognition in videos
Knowledge Transfer for Scene-specific Motion Prediction
Modelling Temporal Information Using Discrete Fourier Transform for Video Classification
Persistent Homology of Attractors For Action Recognition
Conformal Predictors for Compound Activity Prediction
Learning zeroth class dictionary for human action recognition
Pose for Action - Action for Pose
Support Vector Machines with Time Series Distance Kernels for Action Classification
Synthesized Classifiers for Zero-Shot Learning
Robust Multi-body Feature Tracker: A Segmentation-free Approach
Learning to SMILE(S)
Segmental Spatiotemporal CNNs for Fine-grained Action Segmentation
Comparative Evaluation of Action Recognition Methods via Riemannian Manifolds, Fisher Vectors and GMMs: Ideal and Challenging Conditions
Joint Recognition and Segmentation of Actions via Probabilistic Integration of Spatio-Temporal Fisher Vectors
Learning a Deep Model for Human Action Recognition from Novel Viewpoints
Combining ConvNets with Hand-Crafted Features for Action Recognition Based on an HMM-SVM Classifier
Order-aware Convolutional Pooling for Video Based Action Recognition
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.