Spatiotemporal Dilated Convolution With Uncertain Matching for Video-Based Crowd Estimation
Year: 2021
Type: article
Abstract: In this paper, we propose a novel SpatioTemporal convolutional Dense Network (STDNet) to address the video-based crowd counting problem, which contains the decomposition of 3D convolution and the 3D spatiotemporal dilated dense convolution to alleviate the rapid growth of the model size caused by the Conv3D layer. Moreover, since the dilated convolution extracts the multiscale features, we combine the dilated convolution with the channel attention block to enhance the feature representations. Due to the error that occurs from the difficulty of labeling crowds, especially for videos, imprecise or standard-inconsistent labels may lead to poor convergence for the model. To address this issue, we further propose a new patch-wise regression loss (PRL) to improve the original pixel-wise loss. Experimental results on three video-based benchmarks, i.e., the UCSD, Mall and WorldExpo’10 datasets, show that STDNet outperforms both image- and video-based state-of-the-art methods. The source codes are released athttps://github.com/STDNet/STDNet.
Source: IEEE Transactions on Multimedia
Authors: Yu-Jen Ma, Hong-Han Shuai, Wen-Huang Cheng
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: G175096498, Enhance and Secure User Experience in Digital Finance with Machine Learning and Blockchain Technologies, Deep Sensing and Situational Understanding in Self-Driving Cars( II ), Disinformation Detection, Intervention, and Simulation: towards a Controllable Online Environment for Disinformation Management( I ), Self-Supervised Learning for Object Tracking, Enhance and Secure User Experience in Digital Finance with Machine Learning and Blockchain Technologies( III ), Intelligent Communications and Networking Technologies for Drone-Cells( III )
OpenAlex ID: W3120524222