標題: | ON THE EFFECT OF DATA IMBALANCE FOR MULTI-LABEL PEDESTRIAN ATTRIBUTE RECOGNITION |
作者: | Wang, Tsaipei Shu, Kai-Chen Chang, Chia-Hao Chen, Yi-Fu 資訊工程學系 Department of Computer Science |
關鍵字: | Human attribute recognition;pedestrian attribute recognition;multi-label classification;data imbalance;classification evaluation metrics |
公開日期: | 1-一月-2018 |
摘要: | Pedestrian attribute recognition has many applications in surveillance and attribute based query, tracking, and person re-identification. The recent trend in deep-learning based pedestrian attribute recognition is to use a shared CNN backbone for feature extraction and multiple subsequent branches for the individual branches. While this allows the end-to-end learning to simultaneously recognize multiple attributes, the data imbalance problem of most attributes becomes a challenge that has not been studied sufficiently for this application. This paper presents studies on how the cost adjustment method affects several common evaluation metrics. We also propose a two-stage training procedure, where an additional fine-tuning stage on the classifier layers only with class-balanced data is shown to improve recognition performances. |
URI: | http://dx.doi.org/10.1109/TAAI.2018.00025 http://hdl.handle.net/11536/151039 |
ISSN: | 2376-6816 |
DOI: | 10.1109/TAAI.2018.00025 |
期刊: | 2018 CONFERENCE ON TECHNOLOGIES AND APPLICATIONS OF ARTIFICIAL INTELLIGENCE (TAAI) |
起始頁: | 74 |
結束頁: | 77 |
顯示於類別: | 會議論文 |