Deep Learning Driven Hypergraph Representation for Image-Based Emotion Recognition
文献类型：会议论文
| 作者 | Huang Yuchi; Lu Hanqing ; Huang, Yuchi
|
| 出版日期 | 2016 |
| 会议日期 | November 12–16, 2016 |
| 会议地点 | Tokyo, Japan |
| 关键词 | Emotion Recognition Hypergraph Deep Convolutional Networks |
| 英文摘要 |
In this paper, we proposed a bi-stage framework for imagebased
emotion recognition by combining the advantages of
deep convolutional neural networks (D-CNN) and hypergraphs.
To exploit the representational power of D-CNN,
we remodeled its last hidden feature layer as the `attribute'
layer in which each hidden unit produces probabilities on a
speci c semantic attribute. To describe the high-order relationship
among facial images, each face was assigned to
various hyperedges according to the computed probabilities
on di erent D-CNN attributes. In this way, we tackled the
emotion prediction problem by a transductive learning approach,
which tends to assign the same label to faces that
share many incidental hyperedges (attributes), with the constraints
that predicted labels of training samples should be
similar to their ground truth labels. We compared the proposed
approach to state-of-the-art methods and its e ectiveness
was demonstrated by extensive experimentation. |
| 会议录 | 国际会议
 |
| 源URL | [http://ir.ia.ac.cn/handle/173211/12861]  |
| 专题 | 自动化研究所_模式识别国家重点实验室_图像与视频分析团队 |
| 通讯作者 | Huang, Yuchi |
| 推荐引用方式 GB/T 7714 | Huang Yuchi,Lu Hanqing,Huang, Yuchi. Deep Learning Driven Hypergraph Representation for Image-Based Emotion Recognition[C]. 见:. Tokyo, Japan. November 12–16, 2016. |
入库方式： OAI收割
来源：自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。