Attention Deep Model With Multi-Scale Deep Supervision for Person Re-Identification

Wu, Di; Wang, Chao; Wu, Yong; Wang, Qi-Cong; Huang, De-Shuang

doi:10.1109/tetci.2020.3034606

Cited by 62 publications

(34 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The above attention model has been proven to be an effective method to improve the performance of deep neural networks. Inspired by these works, more scholars [28][29][30][31][32][33][34][35] now use an attention mechanism in person Re-ID tasks. Chen et al [31] propose two different attention branches in person Re-ID tasks to enable the learned feature map to perceive persons and related body parts, respectively.…”

Section: Related Workmentioning

confidence: 99%

Part‐level attention networks for cross‐domain person re‐identification

Qun

Du²,

Ouyang³

et al. 2021

IET Image Processing

View full text Add to dashboard Cite

Person re-identification (Re-ID) is in significant demand for intelligent security and single or multiple-target tracking. However, there are issues in the person Re-ID tasks, such as sharp decline in cross-data sets detection accuracy, poor generalization and cross-domain ability of the model. This work mainly studies the generalization and adaptation of crossdomain person Re-ID models. Different from most existing methods for cross-domain Re-ID tasks, the authors use diversified spatial semantic feature in pixel-level learning in the target domain to improve the generality and adaptability of the model. In the case that no information of the target domain is used during the model training, the trained model is directly tested on the data set of the target domain. It has proven effective to add the attention cascade module into the backbone network combining with the part-level branch. The authors conducted extensive experiments based on the three data sets of Market-1501, DukeMTMC-ReID and MSMT17, resulting in both single-domain and cross-domain tests with an average improvement of Rank1 and mAP values of about 10% compared with Baseline through the authors' proposed method named Part-Level Attention Network.

show abstract

Section: Related Workmentioning

confidence: 99%

Part‐level attention networks for cross‐domain person re‐identification

Qun

Du²,

Ouyang³

et al. 2021

IET Image Processing

View full text Add to dashboard Cite

show abstract

“…Compared with the above methods, the proposed method can realize the autonomous suppression operation of features without additional positioning information. The attention modules are used in the most advanced person re-identification architecture [ 2 , 3 , 15 , 17 ], which assign different weights to the channels and spaces features, allowing the model to pay more attention to the important features of a person and suppress unnecessary interference. The above methods are dedicated to strengthening the encoding of the rich features in the input image, but due to the drastic visual angle changes, the features of the consistent salient regions may introduce errors, that is to say, the discriminative information that can be used to match two people may not be available in all situations.…”

Section: Related Workmentioning

confidence: 99%

“…To improve the performance of the model, a variety of research methods were proposed in recent years, which are roughly divided into the following categories: (1) Global feature representation learning—the literature [ 1 , 2 , 3 ] fully excavated the shallow high-resolution and deep high-semantic information and used global average pooling in different stages of the residual network to generate the embedding of each stage and obtained global information at different depths of the network. Qian et al [ 4 ] designed a multi-scale deep learning model and used an adaptive method to find features of an appropriate scale for person retrieval; (2) Local feature representation learning—this commits to discovering more fine-grained visual information distributed throughout the human body.…”

Section: Introductionmentioning

confidence: 99%

“…Part-based Convolutional Baseline (PCB) [ 9 ] uses a refined part pooling strategy to make the feature closer to the real region; (3) Optimization strategy—some works attempted to achieve better person features with the modification of the backbone architecture, data enhancement, and regularization. The literature [ 2 , 3 ] changed the down-sampling convolutional stripe in the final stage of ResNet50 to 1. OSNet [ 10 ] introduced the residual block composed of multiple convolution streams to realize multi-scale feature learning.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Self-Erasing Network for Person Re-Identification

Fan

Yang

Zhang

et al. 2021

Sensors

View full text Add to dashboard Cite

Person re-identification (ReID) plays an important role in intelligent surveillance and receives widespread attention from academics and the industry. Due to extreme changes in viewing angles, some discriminative local regions are suppressed. In addition, the data with similar backgrounds collected by a fixed viewing angle camera will also affect the model’s ability to distinguish a person. Therefore, we need to discover more fine-grained information to form the overall characteristics of each identity. The proposed self-erasing network structure composed of three branches benefits the extraction of global information, the suppression of background noise and the mining of local information. The two self-erasing strategies that we proposed encourage the network to focus on foreground information and strengthen the model’s ability to encode weak features so as to form more effective and richer visual cues of a person. Extensive experiments show that the proposed method is competitive with the advanced methods and achieves state-of-the-art performance on DukeMTMC-ReID and CUHK-03(D) datasets. Furthermore, it can be seen from the activation map that the proposed method is beneficial to spread the attention to the whole body. Both metrics and the activation map validate the effectiveness of our proposed method.

show abstract

“…In the field of person re-identification and object detection, the use of deep supervision can effectively improve the network performance [26,27]. When applying this idea with multi-resolution learning to the segmentation task, it is important to achieve balanced loss by considering different contribution of each resolution tasks [28].…”

Section: Introductionmentioning

confidence: 99%

Multi-Resolution Supervision Network with an Adaptive Weighted Loss for Desert Segmentation

et al. 2021

View full text Add to dashboard Cite

Desert segmentation of remote sensing images is the basis of analysis of desert area. Desert images are usually characterized by large image size, large-scale change, and irregular location distribution of surface objects. The multi-scale fusion method is widely used in the existing deep learning segmentation models to solve the above problems. Based on the idea of multi-scale feature extraction, this paper took the segmentation results of each scale as an independent optimization task and proposed a multi-resolution supervision network (MrsSeg) to further improve the desert segmentation result. Due to the different optimization difficulty of each branch task, we also proposed an auxiliary adaptive weighted loss function (AWL) to automatically optimize the training process. MrsSeg first used a lightweight backbone to extract different-resolution features, then adopted a multi-resolution fusion module to fuse the local information and global information, and finally, a multi-level fusion decoder was used to aggregate and merge the features at different levels to get the desert segmentation result. In this method, each branch loss was treated as an independent task, AWL was proposed to calculate and adjust the weight of each branch. By giving priority to the easy tasks, the improved loss function could effectively improve the convergence speed of the model and the desert segmentation result. The experimental results showed that MrsSeg-AWL effectively improved the learning ability of the model and has faster convergence speed, lower parameter complexity, and more accurate segmentation results.

show abstract

Attention Deep Model With Multi-Scale Deep Supervision for Person Re-Identification

Cited by 62 publications

References 53 publications

Part‐level attention networks for cross‐domain person re‐identification

Part‐level attention networks for cross‐domain person re‐identification

Self-Erasing Network for Person Re-Identification

Multi-Resolution Supervision Network with an Adaptive Weighted Loss for Desert Segmentation

Contact Info

Product

Resources

About