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.
JOLO-GCN: Mining Joint-Centered Light-Weight Information for Skeleton-Based Action Recognition
Recovering and Simulating Pedestrians in the Wild
Prototypical Contrast and Reverse Prediction: Unsupervised Skeleton Based Action Recognition
Towards Zero-Shot Learning with Fewer Seen Class Examples
Transductive Zero-Shot Learning using Cross-Modal CycleGAN
Zero-shot Relation Classification from Side Information
Progressive Spatio-Temporal Graph Convolutional Network for Skeleton-Based Human Action Recognition
Transferred Fusion Learning using Skipped Networks
Multi-Temporal Convolutions for Human Action Recognition in Videos
On the spatial attention in Spatio-Temporal Graph Convolutional Networks for skeleton-based human action recognition
Motion Prediction on Self-driving Cars: A Review
Uncertainty-Aware Vehicle Orientation Estimation for Joint Detection-Prediction Models
Ellipse Loss for Scene-Compliant Motion Prediction
Beelines: Motion Prediction Metrics for Self-Driving Safety and Comfort
Temporally-Continuous Probabilistic Prediction using Polynomial Trajectory Parameterization
Bubblenet: A Disperse Recurrent Structure To Recognize Activities
SAR-NAS: Skeleton-based Action Recognition via Neural Architecture Searching
CNN based Multistage Gated Average Fusion (MGAF) for Human Action Recognition Using Depth and Inertial Sensors
ElderSim: A Synthetic Data Generation Platform for Human Action Recognition in Eldercare Applications
Multi-Class Zero-Shot Learning for Artistic Material Recognition
Temporal Attention-Augmented Graph Convolutional Network for Efficient Skeleton-Based Human Action Recognition
A Software Architecture for Autonomous Vehicles: Team LRM-B Entry in the First CARLA Autonomous Driving Challenge
Learning Graph-Based Priors for Generalized Zero-Shot Learning
Two-Stream Consensus Network for Weakly-Supervised Temporal Action Localization
Deep Analysis of CNN-based Spatio-temporal Representations for Action Recognition
PRANK: motion Prediction based on RANKing
Zero-Shot Learning from scratch (ZFS): leveraging local compositional representations
Stronger, Faster and More Explainable: A Graph Convolutional Baseline for Skeleton-based Action Recognition
Depth Guided Adaptive Meta-Fusion Network for Few-shot Video Recognition
Unsupervised Domain Adaptation for Spatio-Temporal Action Localization
A Grid-based Representation for Human Action Recognition
Pose And Joint-Aware Action Recognition
Multi-label Few/Zero-shot Learning with Knowledge Aggregated from Multiple Label Graphs
LTN: Long-Term Network for Long-Term Motion Prediction
No Rumours Please! A Multi-Indic-Lingual Approach for COVID Fake-Tweet Detection
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.