Multi-label Contrastive Focal Loss for Pedestrian Attribute Recognition
| Autor: | Shilong Wang, Fan Zhu, Xiaoqiang Zheng, Zhenxia Yu, Lin Chen |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
business.industry
Computer science media_common.quotation_subject Feature extraction Pattern recognition 02 engineering and technology 010501 environmental sciences Complex network 01 natural sciences Convolutional neural network Term (time) Visualization Discriminative model Pattern recognition (psychology) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence business Function (engineering) 0105 earth and related environmental sciences media_common |
| Zdroj: | ICPR |
| DOI: | 10.1109/icpr48806.2021.9411959 |
| Popis: | Pedestrian Attribute Recognition (PAR) has received extensive attention during the past few years. With the advances of deep convolutional neural networks (CNNs), the performance of PAR has been significantly improved. Existing methods tend to acquire attribute-specific features by designing various complex network structures with additional modules. Such additional modules, however, dramatically increase the number of network parameters. Meanwhile, the problems of class imbalance and hard attribute retrieving remain underestimated in PAR. In this paper, we explore the optimization mechanism of the training processing to account for these problems and propose a new loss function called Multi-label Contrastive Focal Loss (MCFL). This proposed MCFL emphasizes the hard and minority attributes by using a separated re-weighting mechanism for different positive and negative classes to alleviate the impact of the imbalance. MCFL is also able to enlarge the gaps between the intra-class of multi-label attributes, to force CNNs to extract more subtle discriminative features. We evaluate the proposed MCFL on three large public pedestrian datasets, including RAP, PA-100K, and PETA. The experimental results indicate that the proposed MCFL with the ResNet-50 backbone is able to outperform other state-of-the-art approaches in term of mean accuracy. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|