Prototype Alignment

Prototype alignment is a technique used to improve the performance of machine learning models, particularly in scenarios with limited labeled data or significant domain shifts. Current research focuses on aligning prototypes (representative features) across different domains or classes, often employing techniques like contrastive learning, prototype rectification, and attention mechanisms within various architectures such as prototypical networks and deep alignment networks. This approach addresses challenges in unsupervised domain adaptation, few-shot learning, and federated learning, leading to improved model generalization and robustness across diverse datasets and applications in fields like computer vision and medical image analysis. The resulting advancements enhance the efficiency and accuracy of model training and deployment in data-scarce or heterogeneous environments.

Papers

August 16, 2024

DPA: Dual Prototypes Alignment for Unsupervised Adaptation of Vision-Language Models
Eman Ali, Sathira Silva, Muhammad Haris Khan
Zero Shot Vision Language Model Pseudo Labeling Noisy Pseudo Label Unsupervised Adaptation Prototype Alignment

August 15, 2024

Addressing Skewed Heterogeneity via Federated Prototype Rectification with Personalization
Shunxin Guo, Hongsong Wang, Shuxia Lin, Zhiqiang Kou, Xin Geng
Personalized Federated Learning Heterogeneous Federated Learning Unknown Heterogeneity Prototype Alignment

June 18, 2024

Cephalometric Landmark Detection across Ages with Prototypical Network
Han Wu, Chong Wang, Lanzhuju Mei, Tong Yang, Min Zhu, Dingggang Shen, Zhiming Cui
Landmark Detection Prototypical Network Different Age Prototype Alignment Cephalometric Landmark

April 11, 2024

PromptSync: Bridging Domain Gaps in Vision-Language Models through Class-Aware Prototype Alignment and Discrimination
Anant Khandelwal
Contrastive Learning Zero Shot Vision Language Model Domain Generalization Domain Gap Group DIscrimination Prototype Alignment

March 11, 2024

Toward Robust Canine Cardiac Diagnosis: Deep Prototype Alignment Network-Based Few-Shot Segmentation in Veterinary Medicine
Jun-Young Oh, In-Gyu Lee, Tae-Eui Kam, Ji-Hoon Jeong
Segmentation Based Approach Image Segmentation Artificial Intelligence Research Shot Segmentation Accurate Segmentation Medical Artificial Intelligence Prototype Alignment Veterinary Medicine

January 29, 2024

Dynamic Prototype Adaptation with Distillation for Few-shot Point Cloud Segmentation
Jie Liu, Wenzhe Yin, Haochen Wang, Yunlu CHen, Jan-Jakob Sonke, Efstratios Gavves
Point Cloud Mutual Distillation Shot 3D Prototype Adaptation Prototype Alignment Prototype Attention

August 16, 2023

GPA-3D: Geometry-aware Prototype Alignment for Unsupervised Domain Adaptive 3D Object Detection from Point Clouds
Ziyu Li, Jingming Guo, Tongtong Cao, Liu Bingbing, Wankou Yang
Point Cloud 3D Detection 3D Detector Prototype Alignment

July 19, 2023

Source-Free Domain Adaptation for Medical Image Segmentation via Prototype-Anchored Feature Alignment and Contrastive Learning
Qinji Yu, Nan Xi, Junsong Yuan, Ziyu Zhou, Kang Dang, Xiaowei Ding
Contrastive Learning Domain Adaptation Medical Image Segmentation Unsupervised Domain Adaptation Source Free Domain Adaptation Cross Modality Prototype Alignment

June 12, 2023

Online Prototype Alignment for Few-shot Policy Transfer
Qi Yi, Rui Zhang, Shaohui Peng, Jiaming Guo, Yunkai Gao, Kaizhao Yuan, Ruizhi Chen, Siming Lan, Xing Hu, Zidong Du, Xishan Zhang, Qi Guo, Yunji Chen
Reinforcement Learning Domain Adaptation Mapping Function Prototype Alignment Shot Transfer Learning

March 28, 2023

Deep Incomplete Multi-view Clustering with Cross-view Partial Sample and Prototype Alignment
Jiaqi Jin, Siwei Wang, Zhibin Dong, Xinwang Liu, En Zhu
Multi View Multi View Clustering Uncertainty Induced Incomplete Alignment Model Incomplete Multi View Clustering Prototype Alignment Multi View Sampling

July 22, 2022

Prototype-Guided Continual Adaptation for Class-Incremental Unsupervised Domain Adaptation
Hongbin Lin, Yifan Zhang, Zhen Qiu, Shuaicheng Niu, Chuang Gan, Yanxia Liu, Mingkui Tan
Domain Adaptation Unlabeled Target Domain Continual Adaptation Domain Incremental Learning Prototype Alignment

June 3, 2022

Mutual- and Self- Prototype Alignment for Semi-supervised Medical Image Segmentation
Zhenxi Zhang, Chunna Tian, Zhicheng Jiao
Long Term Cooperation Semi Supervised Medical Image Segmentation Semi Supervised Segmentation Prototype Alignment

November 24, 2021

APANet: Adaptive Prototypes Alignment Network for Few-Shot Semantic Segmentation
Jiacheng Chen, Bin-Bin Gao, Zongqing Lu, Jing-Hao Xue, Chengjie Wang, Qingmin Liao
Semantic Segmentation Shot Semantic Segmentation Prototype Selection Prototype Alignment Unbiased Segmentation

Prototype Alignment

Papers

DPA: Dual Prototypes Alignment for Unsupervised Adaptation of Vision-Language Models

Addressing Skewed Heterogeneity via Federated Prototype Rectification with Personalization

Cephalometric Landmark Detection across Ages with Prototypical Network

PromptSync: Bridging Domain Gaps in Vision-Language Models through Class-Aware Prototype Alignment and Discrimination

Toward Robust Canine Cardiac Diagnosis: Deep Prototype Alignment Network-Based Few-Shot Segmentation in Veterinary Medicine

Dynamic Prototype Adaptation with Distillation for Few-shot Point Cloud Segmentation

GPA-3D: Geometry-aware Prototype Alignment for Unsupervised Domain Adaptive 3D Object Detection from Point Clouds

Source-Free Domain Adaptation for Medical Image Segmentation via Prototype-Anchored Feature Alignment and Contrastive Learning

Online Prototype Alignment for Few-shot Policy Transfer

Deep Incomplete Multi-view Clustering with Cross-view Partial Sample and Prototype Alignment

Prototype-Guided Continual Adaptation for Class-Incremental Unsupervised Domain Adaptation

Mutual- and Self- Prototype Alignment for Semi-supervised Medical Image Segmentation

APANet: Adaptive Prototypes Alignment Network for Few-Shot Semantic Segmentation