Visual search and application using deep learning (Age group classification with convolution neural network)
The consistency of processing input facial images to automatically estimating human age has generally been found to be poor. Facial image classification is an approach to classify face images into a few predefined age groups. Predicting age group from face images acquired in unconstrained conditions...
Saved in:
主要作者: | |
---|---|
其他作者: | |
格式: | Final Year Project |
語言: | English |
出版: |
2018
|
主題: | |
在線閱讀: | http://hdl.handle.net/10356/74924 |
標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
機構: | Nanyang Technological University |
語言: | English |
總結: | The consistency of processing input facial images to automatically estimating human age has generally been found to be poor. Facial image classification is an approach to classify face images into a few predefined age groups. Predicting age group from face images acquired in unconstrained conditions have been a challenging and important task in many modern applications. With the rise of social media sites, this has become relevant to an increasing amount of applications. Nevertheless, performance of existing methods with manually-designed features on in-the-wild benchmarks are still lacking in this area, as compared to the improving results in performance reported for the related task of facial recognition. In this paper, a deep CNN model that was trained for face recognition task is used to estimate the age information on the IMDB-WIKI database. To this end, a proposed simple CNN architecture can be used even when the amount of learning data is limited. Our experiments illustrate the effectiveness of different models for age group estimation in the wild. Further studies will be evaluated on the method used on the recent IMDB-WIKI benchmark for age group classification and compared it to other state of the art methods. |
---|