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.
Ordered Pooling of Optical Flow Sequences for Action Recognition
AENet: Learning Deep Audio Features for Video Analysis
Action Recognition Based on Joint Trajectory Maps with Convolutional Neural Networks
Human Action Attribute Learning From Video Data Using Low-Rank Representations
DeMIAN: Deep Modality Invariant Adversarial Network
Efficient Action Detection in Untrimmed Videos via Multi-Task Learning
Deep Motion Features for Visual Tracking
Dynamic Action Recognition: A convolutional neural network model for temporally organized joint location data
Asynchronous Temporal Fields for Action Recognition
Deep Learning on Lie Groups for Skeleton-based Action Recognition
Tinkering Under the Hood: Interactive Zero-Shot Learning with Net Surgery
Jointly learning heterogeneous features for rgb-d activity recognition
Single Image Action Recognition using Semantic Body Part Actions
Scale Coding Bag of Deep Features for Human Attribute and Action Recognition
ActionFlowNet: Learning Motion Representation for Action Recognition
FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks
Zero-Shot Learning posed as a Missing Data Problem
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
Action Recognition with Dynamic Image Networks
Procedural Generation of Videos to Train Deep Action Recognition Networks
Deep Quantization: Encoding Convolutional Activations with Deep Generative Model
Unifying Multi-Domain Multi-Task Learning: Tensor and Neural Network Perspectives
Action Recognition Based on Optimal Joint Selection and Discriminative Depth Descriptor
Multi-Task Zero-Shot Action Recognition with Prioritised Data Augmentation
AdaScan: Adaptive Scan Pooling in Deep Convolutional Neural Networks for Human Action Recognition in Videos
Learning Joint Feature Adaptation for Zero-Shot Recognition
Adaptive Down-Sampling and Dimension Reduction in Time Elastic Kernel Machines for Efficient Recognition of Isolated Gestures
Learning Multi-level Features For Sensor-based Human Action Recognition
Self-Supervised Video Representation Learning With Odd-One-Out Networks
An End-to-End Spatio-Temporal Attention Model for Human Action Recognition from Skeleton Data
Deep Action- and Context-Aware Sequence Learning for Activity Recognition and Anticipation
Learning To Score Olympic Events
Joint Network based Attention for Action Recognition
Learning long-term dependencies for action recognition with a biologically-inspired deep network
Temporal Convolutional Networks for Action Segmentation and Detection
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.