Atrous convolutions spatial pyramid network for crowd counting and density estimation

Ma, Junjie; Dai, Yaping; Tan, Yap Peng

doi:10.1016/j.neucom.2019.03.065

Cited by 43 publications

(29 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By setting the size of the dilation rate, CSRNet could provide different receptive fields to extract features of different scales of an image and successfully achieve the technology transformation from the image segmentation field to the crowd counting. On this basis, Ma et al [20] proposed an atrous convolutions spatial pyramid network (ACSPNet), in which the convolutional blocks with different void rates integrate multiscale information and range through jump connection to improve scale perception. Although the application of dilated convolution greatly simplifies the network, it cannot reflect his superiority when there are objects with a lower resolution in the image.…”

Section: Related Workmentioning

confidence: 99%

Crowd Counting Based on Multiresolution Density Map and Parallel Dilated Convolution

Tang

Zhou

et al. 2021

Scientific Programming

View full text Add to dashboard Cite

The current crowd counting tasks rely on a fully convolutional network to generate a density map that can achieve good performance. However, due to the crowd occlusion and perspective distortion in the image, the directly generated density map usually neglects the scale information and spatial contact information. To solve it, we proposed MDPDNet (Multiresolution Density maps and Parallel Dilated convolutions’ Network) to reduce the influence of occlusion and distortion on crowd estimation. This network is composed of two modules: (1) the parallel dilated convolution module (PDM) that combines three dilated convolutions in parallel to obtain the deep features on the larger receptive field with fewer parameters while reducing the loss of multiscale information; (2) the multiresolution density map module (MDM) that contains three-branch networks for extracting spatial contact information on three different low-resolution density maps as the feature input of the final crowd density map. Experiments show that MDPDNet achieved excellent results on three mainstream datasets (ShanghaiTech, UCF_CC_50, and UCF-QNRF).

show abstract

Section: Related Workmentioning

confidence: 99%

Crowd Counting Based on Multiresolution Density Map and Parallel Dilated Convolution

Tang

Zhou

et al. 2021

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…Crowd analysis is a popular task in computer vision [1,2,3,4], which focuses on understand the still or video crowd scenes at a high level. In the field of crowd analysis, crowd counting [5,6,7,8] is an essential branch, which focuses on predicting the number of people or estimating the density maps for crowd scenes. Accurate crowd counting is important to urban safety, public design, space management and so on.…”

Section: Introductionmentioning

confidence: 99%

SCAR: Spatial-/channel-wise attention regression networks for crowd counting

2019

View full text Add to dashboard Cite

Recently, crowd counting is a hot topic in crowd analysis. Many CNNbased counting algorithms attain good performance. However, these methods only focus on the local appearance features of crowd scenes but ignore the large-range pixel-wise contextual and crowd attention information. To remedy the above problems, in this paper, we introduce the Spatial-/Channelwise Attention Models into the traditional Regression CNN to estimate the density map, which is named as "SCAR". It consists of two modules, namely Spatial-wise Attention Model (SAM) and Channel-wise Attention Model (CAM). The former can encode the pixel-wise context of the entire image to more accurately predict density maps at the pixel level. The latter attempts to extract more discriminative features among different channels, which aids model to pay attention to the head region, the core of crowd scenes. Intuitively, CAM alleviates the mistaken estimation for background regions. Finally, two types of attention information and traditional CNN's feature maps are integrated by a concatenation operation. Furthermore, the extensive experiments are conducted on four popular datasets, Shanghai Tech Part A/B, GCC, and UCF CC 50 Dataset. The results show that the proposed method achieves state-of-the-art results.

show abstract

“…Deep learning technology can continuously improve its performance with the increase of data scale, while in traditional machine learning algorithms it is difficult to use massive data to continuously improve their performance. DL has shown great application prospects in many research fields [ 14 , 15 , 16 ]. For example, DL’s performance in image classification, object detection, semantic segmentation and other tasks in the field of computer vision greatly surpasses traditional methods, while in the field of natural language processing has played an irreplaceable role in the research of tasks such as speech recognition, machine translation, and dialogue systems, and has also achieved breakthrough results in autonomous driving, medical health, and fault diagnosis.…”

Section: Introductionmentioning

confidence: 99%

Fault Diagnosis of Rolling Bearings Based on a Residual Dilated Pyramid Network and Full Convolutional Denoising Autoencoder

Shi

Chen

et al. 2020

Sensors

View full text Add to dashboard Cite

Intelligent fault diagnosis algorithm for rolling bearings has received increasing attention. However, in actual industrial environments, most rolling bearings work under severe working conditions of variable speed and strong noise, which makes the performance of many intelligent fault diagnosis methods deteriorate sharply. In this regard, this paper proposes a new intelligent diagnosis algorithm for rolling bearing faults based on a residual dilated pyramid network and full convolutional denoising autoencoder (RDPN-FCDAE). First, a continuous wavelet transform (CWT) is used to convert original vibration signals into time-frequency images. Secondly, a deep two-stage RDPN-FCDAE model is constructed, which is divided into three parts: encoding network, decoding network and classification network. In order to obtain efficient expression of data denoising feature of encoding network, time-frequency images are first input into the encoding-decoding network for unsupervised pre-training. Then pre-trained coding network and classification network are combined into residual dilated pyramid full convolutional network (RDPFCN) for parameter fine-tuning and testing. The proposed method is applied to bearing vibration datasets of test rig with different speeds and noise modes. Compared with representative machine learning and deep learning method, the results show that the algorithm proposed is superior to other methods in diagnostic accuracy, noise robustness and feature segmentation ability.

show abstract

Atrous convolutions spatial pyramid network for crowd counting and density estimation

Cited by 43 publications

References 17 publications

Crowd Counting Based on Multiresolution Density Map and Parallel Dilated Convolution

Crowd Counting Based on Multiresolution Density Map and Parallel Dilated Convolution

SCAR: Spatial-/channel-wise attention regression networks for crowd counting

Fault Diagnosis of Rolling Bearings Based on a Residual Dilated Pyramid Network and Full Convolutional Denoising Autoencoder

Contact Info

Product

Resources

About