An Efficient Multi-Scale Focusing Attention Network for Person Re-Identification
| Autor: | Jihui Xu, Xiaoyu Hou, Wei Huang, Yongying Li, Huaiyu Xu, Kunlin Zhang, Ruidan Su |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Computational complexity theory
Computer science 02 engineering and technology lcsh:Technology 01 natural sciences lcsh:Chemistry lightweight network 0103 physical sciences Convergence (routing) 0202 electrical engineering electronic engineering information engineering General Materials Science Latency (engineering) lcsh:QH301-705.5 Instrumentation Block (data storage) 010302 applied physics Fluid Flow and Transfer Processes multi-scale feature learning person ReID lcsh:T Process Chemistry and Technology General Engineering Grid softmax and TriHard loss lcsh:QC1-999 Computer Science Applications lcsh:Biology (General) lcsh:QD1-999 Computer engineering lcsh:TA1-2040 Softmax function Focusing attention 020201 artificial intelligence & image processing lcsh:Engineering (General). Civil engineering (General) attention mechanism lcsh:Physics Communication channel |
| Zdroj: | Applied Sciences Volume 11 Issue 5 Applied Sciences, Vol 11, Iss 2010, p 2010 (2021) |
| ISSN: | 2076-3417 |
| DOI: | 10.3390/app11052010 |
| Popis: | The multi-scale lightweight network and attention mechanism recently attracted attention in person re-identification (ReID) as it is capable of improving the model’s ability to process information with low computational cost. However, state-of-the-art methods mostly concentrate on the spatial attention and big block channel attention model with high computational complexity while rarely investigate the inside block attention with the lightweight network, which cannot meet the requirements of high efficiency and low latency in the actual ReID system. In this paper, a novel lightweight person ReID model is designed firstly, called Multi-Scale Focusing Attention Network (MSFANet), to capture robust and elaborate multi-scale ReID features, which have fewer float-computing and higher performance. MSFANet is achieved by designing a multi-branch depthwise separable convolution module, combining with an inside block attention module, to extract and fuse multi-scale features independently. In addition, we design a multi-stage backbone with the ‘1-2-3’ form, which can significantly reduce computational cost. Furthermore, the MSFANet is exceptionally lightweight and can be embedded in a ReID framework flexibly. Secondly, an efficient loss function combining softmax loss and TriHard loss, based on the proposed optimal data augmentation method, is designed for faster convergence and better model generalization ability. Finally, the experimental results on two big ReID datasets (Market1501 and DukeMTMC) and two small ReID datasets (VIPeR, GRID) show that the proposed MSFANet achieves the best mAP performance and the lowest computational complexity compared with state-of-the-art methods, which are increasing by 2.3% and decreasing by 18.2%, respectively. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|