Affinity Learning

Affinity learning in computer vision focuses on leveraging relationships between data points (pixels, points in a point cloud, etc.) to improve tasks like semantic segmentation and instance segmentation, particularly in weakly supervised settings. Current research emphasizes developing efficient multitask learning architectures, often incorporating transformers or convolutional neural networks with novel modules to learn and refine these affinities, improving performance by integrating local and global context, and handling challenges like long-tailed distributions and varying scales. These advancements are significantly impacting various applications, including medical image analysis and scene understanding, by enabling more accurate and efficient segmentation from limited labeled data.

Papers

October 31, 2024

Localization, balance and affinity: a stronger multifaceted collaborative salient object detector in remote sensing images
Yakun Xie, Suning Liu, Hongyu Chen, Shaohan Cao, Huixin Zhang, Dejun Feng, Qian Wan, Jun Zhu, Qing Zhu
Object Detector Localization Focus Remote Sensing Image SALient Object Detection Balancing Strategy Affinity Learning

March 2, 2024

Auxiliary Tasks Enhanced Dual-affinity Learning for Weakly Supervised Semantic Segmentation
Lian Xu, Mohammed Bennamoun, Farid Boussaid, Wanli Ouyang, Ferdous Sohel, Dan Xu
Semantic Segmentation Saliency Map Weakly Supervised Semantic Segmentation Saliency Network Pseudo Segmentation Affinity Learning

January 20, 2024

Cross-Task Affinity Learning for Multitask Dense Scene Predictions
Dimitrios Sinodinos, Narges Armanfard
Multitask Learning Task Affinity Affinity Learning

December 27, 2023

Multi-modality Affinity Inference for Weakly Supervised 3D Semantic Segmentation
Xiawei Li, Qingyuan Xu, Jing Zhang, Tianyi Zhang, Qian Yu, Lu Sheng, Dong Xu
Point Cloud Segmentation Scene Level Label Weakly Supervised 3D Affinity Learning

May 30, 2023

Scale-aware Super-resolution Network with Dual Affinity Learning for Lesion Segmentation from Medical Images
Yanwen Li, Luyang Luo, Huangjing Lin, Pheng-Ann Heng, Hao Chen
Medical Image Lesion Segmentation Affinity Learning

April 12, 2022

Open-World Instance Segmentation: Exploiting Pseudo Ground Truth From Learned Pairwise Affinity
Weiyao Wang, Matt Feiszli, Heng Wang, Jitendra Malik, Du Tran
Ground Truth Mask Prediction Instance Mask Grouping Network Open World Instance Segmentation Affinity Learning

March 5, 2022

Learning Affinity from Attention: End-to-End Weakly-Supervised Semantic Segmentation with Transformers
Lixiang Ru, Yibing Zhan, Baosheng Yu, Bo Du
Convolutional Neural Network Semantic Segmentation Transformer Megatron Decepticons End to End Human Attention Multi Head Self Attention Image Level Label Semantic Affinity Affinity Learning

Affinity Learning

Papers

Localization, balance and affinity: a stronger multifaceted collaborative salient object detector in remote sensing images

Auxiliary Tasks Enhanced Dual-affinity Learning for Weakly Supervised Semantic Segmentation

Cross-Task Affinity Learning for Multitask Dense Scene Predictions

Multi-modality Affinity Inference for Weakly Supervised 3D Semantic Segmentation

Scale-aware Super-resolution Network with Dual Affinity Learning for Lesion Segmentation from Medical Images

Open-World Instance Segmentation: Exploiting Pseudo Ground Truth From Learned Pairwise Affinity

Learning Affinity from Attention: End-to-End Weakly-Supervised Semantic Segmentation with Transformers