经典论文阅读——DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations (CVPR 2
2017-08-09 10:01
696 查看
DeepFashion: Powering Robust Clothes Recognition and
Retrieval with Rich Annotations (CVPR 2016)
link:http://mmlab.ie.cuhk.edu.hk/projects/DeepFashion.html代码实现:https://github.com/liuziwei7/fashion-detection
介绍了衣服识别和搜索,同样是与实例搜索相关的任务,来自于香港中文大学Ziwei Liu等人的工作。首先,本篇文章介绍了一个名为DeepFashion的衣服数据库。该数据库包含超过800K张的衣服图片,50个细粒度类别和1000个属性,并还额外提供衣服的关键点和跨姿态/跨领域的衣服对关系(cross-pose/cross-domain
pair correspondences)
然后为了说明该数据库的效果,作者提出了一种新颖的深度学习网络,FashionNet——通过联合预测衣服的关键点和属性,学习得到更具区分性的特征。该网络的总体框架如下所示:
FashionNet的前向计算过程总共分为三个阶段:第一个阶段,将一张衣服图片输入到网络中的蓝色分支,去预测衣服的关键点是否可见和位置。第二个阶段,根据在上一步预测的关键点位置,关键点池化层(landmark pooling layer)得到衣服的局部特征。第三个阶段,将“fc6
global”层的全局特征和“fc6 local”的局部特征拼接在一起组成“fc7_fusion”,作为最终的图像特征。FashionNet引入了四种损失函数,并采用一种迭代训练的方式去优化。这些损失分别为:回归损失对应于关键点定位,softmax损失对应于关键点是否可见和衣服类别,交叉熵损失函数对应属性预测和三元组损失函数对应于衣服之间的相似度学习。作者分别从衣服分类,属性预测和衣服搜索这三个方面,将FashionNet与其他方法相比较,都取得了明显更好的效果。
相关文章推荐
- DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations论文阅读
- 论文阅读理解 - DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations
- DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations – CVPR 2016
- 论文理解:DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations
- DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations – CVPR 2016
- DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations – CVPR 2016
- DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations
- Fast and Accurate Entity Recognition with Iterated Dilated Convolutions 论文阅读
- 论文阅读(Lukas Neuman——【ICDAR2015】Efficient Scene Text Localization and Recognition with Local Character Refinement)
- Extracting and Composing Robust Features with Denoising Autoencoders(经典文章阅读)
- 论文阅读:CVPR 2015 FaceNet: A Unified Embedding for Face Recognition and Clustering
- 【论文阅读笔记】CVPR2015-Long-term Recurrent Convolutional Networks for Visual Recognition and Description
- [论文阅读]词汇树 Scalable Recognition with a Vocabulary Tree
- 论文阅读:Deep Filter Banks for Texture Recognition and Segmentation
- 论文阅读笔记-ClusType: Effective Entity Recognition and Typing by Relation Phrase-Based Clustering
- 论文阅读(Xiang Bai——【TIP2014】A Unified Framework for Multi-Oriented Text Detection and Recognition)
- 论文阅读(Lukas Neumann——【ICCV2017】Deep TextSpotter_An End-to-End Trainable Scene Text Localization and Recognition Framework)
- 论文阅读:ROBUST AND FULLY AUTOMATED SEGMENTATION OF MANDIBLE FROM CT SCANS
- 阅读小结:Fine-Grained Recognition with Automatic and Efficient Part Attention
- 论文阅读笔记:R-CNN:Rich feature hierarchies for accurate object detection and semantic segmentation