READING NOTE: Learning Spatial Regularization with Image-level Supervisions for Multi-label ...
来源:互联网 发布:下列不是数据库管理软 编辑:程序博客网 时间:2024/06/13 03:07
TITLE: Learning Spatial Regularization with Image-level Supervisions for Multi-label Image Classification
AUTHOR: Feng Zhu, Hongsheng Li, Wanli Ouyang, Nenghai Yu, Xiaogang Wang
ASSOCIATION: University of Science and Technology of China, University of Sydney, The Chinese University of Hong Kong
FROM: arXiv:1702.05891
CONTRIBUTIONS
- An end-to-end deep neural network for multi-label image classification is proposed, which exploits both semantic and spatial relations of labels by training learnable convolutions on the attention maps of labels. Such relations are learned with only image-level supervisions. Investigation and visualization of learned models demonstrate that our model can effectively capture semantic and spatial relations of labels.
- The proposed algorithm has great generalization capability and works well on data with different types of labels.
METHOD
The proposed Spatial Regularization Net (SRN) takes visual features from the main net as inputs and learns to regularize spatial relations between labels. Such relations are exploited based on the learned attention maps for the multiple labels. Label confidences from both main net and SRN are aggregated to generate final classification confidences. The whole network is a unified framework and is trained in an end-to-end manner.
The scheme of SRN is illustrated in the following figure.
To train the network,
- Finetune only the main net on the target dataset. Both
fcnn andfcls are learned with cross-entropy loss for classification. - Fix
fcnn andfcls . Trainfatt andconv1 with cross-entropy loss for classification. - Train
fsr with cross-entropy loss for classification by fixing all other sub-networks. - The whole network is jointly finetuned with joint loss.
The main network follows the structure of ResNet-101. And it is finetuned on the target dataset. The output of Attention Map and Confidence Map has
- READING NOTE: Learning Spatial Regularization with Image-level Supervisions for Multi-label ...
- Learning Spatial Regularization with Image-level Supervisions for Multi-label Image Classification
- KNN for Multi-label learning
- Multi-label learning for BP
- Barcelona Dataset for Multi-label Image Annotation
- READING NOTE: Learning Deconvolution Network for Semantic Segmentation
- READING NOTE: What makes ImageNet good for transfer learning?
- Instance-Aware Hashing for Multi-Label Image Retrieval
- READING NOTE: Deep Image Matting
- CalTech machine learning, video 12 note(regularization)
- Multi-label learning 第一周
- Multi-label learning (转)
- Reading Note: ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression
- [Paper note] Joint Learning of Single-image and Cross-image Representations for Person Re-id.
- 论文笔记 | CNN-RNN:A Unified Framework for Multi-label Image Classification
- 多标签图像分类--HCP: A Flexible CNN Framework for Multi-Label Image Classification
- HCP: A Flexible CNN Framework for Multi-label Image Classification论文学习
- Joint Patch and Multi-label Learning for Facial Action Unit Detection
- 排序:选择排序(算法)
- 用多线程模拟5000人抢1000张票, 测试volatile, synchronized, vector的效用
- Redis高级特性及应用场景
- Java多线程-线程按序交替
- 使用libmodbus库快速构建上位机(二) 串口modbus上位机
- READING NOTE: Learning Spatial Regularization with Image-level Supervisions for Multi-label ...
- app崩溃的原因 和 提前测试流程/方法 和出现崩溃后怎么定位和处理 总结(持续更新中)
- 常见问题
- iOS中GCD的使用小结
- MySQL(三)
- HTTP Keep-Alive是什么?如何工作?
- MTK flashtool 在14.04上不能使用问题
- Servlet常用接口、作用域及停止方法
- Django开发实践-2