Papers with Code 아카이브
Novel Class Discovery
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 0
- 평가표
- 3
- 보존된 평가 기록
- 3
The goal of Novel Class Discovery (NCD) is to identify new classes in unlabeled data, by exploiting prior knowledge from known classes. In this specific setup, the data is split in two sets. The first is a labeled set containing known classes and the second is an unlabeled set containing unknown classes that must be discovered.
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
DATA: Multi-Disentanglement based Contrastive Learning for Open-World Semi-Supervised Deepfake Attribution
Freeze and Cluster: A Simple Baseline for Rehearsal-Free Continual Category Discovery
Novel Class Discovery for Open Set Raga Classification
AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios
OW-Rep: Open World Object Detection with Instance Representation Learning
Hierarchical novel class discovery for single-cell transcriptomic profiles
Text3DAug -- Prompted Instance Augmentation for LiDAR Perception
NC-NCD: Novel Class Discovery for Node Classification
Dual-level Adaptive Self-Labeling for Novel Class Discovery in Point Cloud Segmentation
Self-Cooperation Knowledge Distillation for Novel Class Discovery
Exclusive Style Removal for Cross Domain Novel Class Discovery
Novel Class Discovery for Ultra-Fine-Grained Visual Categorization
Continual Novel Class Discovery via Feature Enhancement and Adaptation
Uncertainty-guided Open-Set Source-Free Unsupervised Domain Adaptation with Target-private Class Segregation
TV100: A TV Series Dataset that Pre-Trained CLIP Has Not Seen
YOLOOC: YOLO-based Open-Class Incremental Object Detection with Novel Class Discovery
Beyond the Known: Novel Class Discovery for Open-world Graph Learning
OV-Uni3DETR: Towards Unified Open-Vocabulary 3D Object Detection via Cycle-Modality Propagation
Adaptive Discovering and Merging for Incremental Novel Class Discovery
Open-world Machine Learning: A Review and New Outlooks
Seeing Unseen: Discover Novel Biomedical Concepts via Geometry-Constrained Probabilistic Modeling
Debiased Novel Category Discovering and Localization
PANDAS: Prototype-based Novel Class Discovery and Detection
Federated Continual Novel Class Learning
Novel class discovery meets foundation models for 3D semantic segmentation
Enhancing Novel Object Detection via Cooperative Foundational Models
A Practical Approach to Novel Class Discovery in Tabular Data
Towards Novel Class Discovery: A Study in Novel Skin Lesions Clustering
Bridging the Gap: Learning Pace Synchronization for Open-World Semi-Supervised Learning
Reinforcement Learning Based Multi-modal Feature Fusion Network for Novel Class Discovery
MetaGCD: Learning to Continually Learn in Generalized Category Discovery
When and How Does Known Class Help Discover Unknown Ones? Provable Understanding Through Spectral Analysis
Novel Class Discovery for Long-tailed Recognition
Class-relation Knowledge Distillation for Novel Class Discovery
An Interactive Interface for Novel Class Discovery in Tabular Data
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.