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.
Actor and Observer: Joint Modeling of First and Third-Person Videos
Memory Attention Networks for Skeleton-based Action Recognition
View Adaptive Neural Networks for High Performance Skeleton-based Human Action Recognition
Rethinking the Faster R-CNN Architecture for Temporal Action Localization
Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation
PM-GANs: Discriminative Representation Learning for Action Recognition Using Partial-modalities
Complementary Attributes: A New Clue to Zero-Shot Learning
Precise Temporal Action Localization by Evolving Temporal Proposals
A Large-scale Attribute Dataset for Zero-shot Learning
STAIR Actions: A Video Dataset of Everyday Home Actions
Zero-Shot Object Detection
Constrained Convolutional Neural Networks: A New Approach Towards General Purpose Image Manipulation Detection
Probabilistic Prediction of Vehicle Semantic Intention and Motion
Audio-Visual Scene Analysis with Self-Supervised Multisensory Features
Guess Where? Actor-Supervision for Spatiotemporal Action Localization
Hierarchical Novelty Detection for Visual Object Recognition
Transductive Unbiased Embedding for Zero-Shot Learning
DIY Human Action Data Set Generation
Attributes as Operators: Factorizing Unseen Attribute-Object Compositions
Non-Linear Temporal Subspace Representations for Activity Recognition
Video Representation Learning Using Discriminative Pooling
Adversarial Framework for Unsupervised Learning of Motion Dynamics in Videos
FaceForensics: A Large-scale Video Dataset for Forgery Detection in Human Faces
Towards Universal Representation for Unseen Action Recognition
Learning and Recognizing Human Action from Skeleton Movement with Deep Residual Neural Networks
Exploiting deep residual networks for human action recognition from skeletal data
Zero-shot Recognition via Semantic Embeddings and Knowledge Graphs
T-RECS: Training for Rate-Invariant Embeddings by Controlling Speed for Action Recognition
Deja Vu: Motion Prediction in Static Images
Featureless: Bypassing feature extraction in action categorization
Attention-based Temporal Weighted Convolutional Neural Network for Action Recognition
Discriminative Learning of Latent Features for Zero-Shot Recognition
Depth Pooling Based Large-scale 3D Action Recognition with Convolutional Neural Networks
Zero-Shot Object Detection: Learning to Simultaneously Recognize and Localize Novel Concepts
Deep Multiple Instance Learning for Zero-shot Image Tagging
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.