Matches in SemOpenAlex for { <https://semopenalex.org/work/W4362588015> ?p ?o ?g. }
- W4362588015 endingPage "873" @default.
- W4362588015 startingPage "859" @default.
- W4362588015 abstract "Abstract Crowd counting provides an important foundation for public security and urban management. Due to the existence of small targets and large density variations in crowd images, crowd counting is a challenging task. Mainstream methods usually apply convolution neural networks (CNNs) to regress a density map, which requires annotations of individual persons and counts. Weakly-supervised methods can avoid detailed labeling and only require counts as annotations of images, but existing methods fail to achieve satisfactory performance because a global perspective field and multi-level information are usually ignored. We propose a weakly-supervised method, DTCC, which effectively combines multi-level dilated convolution and transformer methods to realize end-to-end crowd counting. Its main components include a recursive swin transformer and a multi-level dilated convolution regression head. The recursive swin transformer combines a pyramid visual transformer with a fine-tuned recursive pyramid structure to capture deep multi-level crowd features, including global features. The multi-level dilated convolution regression head includes multi-level dilated convolution and a linear regression head for the feature extraction module. This module can capture both low- and high-level features simultaneously to enhance the receptive field. In addition, two regression head fusion mechanisms realize dynamic and mean fusion counting. Experiments on four well-known benchmark crowd counting datasets (UCF_CC_50, ShanghaiTech, UCF_QNRF, and JHU-Crowd++) show that DTCC achieves results superior to other weakly-supervised methods and comparable to fully-supervised methods." @default.
- W4362588015 created "2023-04-06" @default.
- W4362588015 creator A5001796793 @default.
- W4362588015 creator A5027575021 @default.
- W4362588015 creator A5050137000 @default.
- W4362588015 creator A5054565575 @default.
- W4362588015 creator A5064222660 @default.
- W4362588015 date "2023-04-02" @default.
- W4362588015 modified "2023-10-17" @default.
- W4362588015 title "DTCC: Multi-level dilated convolution with transformer for weakly-supervised crowd counting" @default.
- W4362588015 cites W1536680647 @default.
- W4362588015 cites W1978232622 @default.
- W4362588015 cites W1987118352 @default.
- W4362588015 cites W2058907003 @default.
- W4362588015 cites W2072232009 @default.
- W4362588015 cites W2120815373 @default.
- W4362588015 cites W2463631526 @default.
- W4362588015 cites W2514654788 @default.
- W4362588015 cites W2520826941 @default.
- W4362588015 cites W2741077351 @default.
- W4362588015 cites W2798490576 @default.
- W4362588015 cites W2886443245 @default.
- W4362588015 cites W2895051362 @default.
- W4362588015 cites W2914913933 @default.
- W4362588015 cites W2962720716 @default.
- W4362588015 cites W2963035940 @default.
- W4362588015 cites W2963693541 @default.
- W4362588015 cites W2964209782 @default.
- W4362588015 cites W2967069910 @default.
- W4362588015 cites W2967776630 @default.
- W4362588015 cites W2982007926 @default.
- W4362588015 cites W2982014038 @default.
- W4362588015 cites W2996703886 @default.
- W4362588015 cites W3004672782 @default.
- W4362588015 cites W3035193053 @default.
- W4362588015 cites W3035307763 @default.
- W4362588015 cites W3081099313 @default.
- W4362588015 cites W3096609285 @default.
- W4362588015 cites W3097407159 @default.
- W4362588015 cites W3138516171 @default.
- W4362588015 cites W3151130473 @default.
- W4362588015 cites W3175630421 @default.
- W4362588015 cites W3175725565 @default.
- W4362588015 cites W3176047859 @default.
- W4362588015 cites W3176458063 @default.
- W4362588015 cites W3189653508 @default.
- W4362588015 cites W3203845557 @default.
- W4362588015 cites W3206836360 @default.
- W4362588015 cites W3214228243 @default.
- W4362588015 cites W4214665794 @default.
- W4362588015 cites W4225264236 @default.
- W4362588015 cites W4312613051 @default.
- W4362588015 cites W4377101865 @default.
- W4362588015 cites W4385257175 @default.
- W4362588015 doi "https://doi.org/10.1007/s41095-022-0313-5" @default.
- W4362588015 hasPublicationYear "2023" @default.
- W4362588015 type Work @default.
- W4362588015 citedByCount "0" @default.
- W4362588015 crossrefType "journal-article" @default.
- W4362588015 hasAuthorship W4362588015A5001796793 @default.
- W4362588015 hasAuthorship W4362588015A5027575021 @default.
- W4362588015 hasAuthorship W4362588015A5050137000 @default.
- W4362588015 hasAuthorship W4362588015A5054565575 @default.
- W4362588015 hasAuthorship W4362588015A5064222660 @default.
- W4362588015 hasBestOaLocation W43625880151 @default.
- W4362588015 hasConcept C105795698 @default.
- W4362588015 hasConcept C153180895 @default.
- W4362588015 hasConcept C154945302 @default.
- W4362588015 hasConcept C31972630 @default.
- W4362588015 hasConcept C33923547 @default.
- W4362588015 hasConcept C41008148 @default.
- W4362588015 hasConcept C45347329 @default.
- W4362588015 hasConcept C50644808 @default.
- W4362588015 hasConcept C52622490 @default.
- W4362588015 hasConcept C83546350 @default.
- W4362588015 hasConceptScore W4362588015C105795698 @default.
- W4362588015 hasConceptScore W4362588015C153180895 @default.
- W4362588015 hasConceptScore W4362588015C154945302 @default.
- W4362588015 hasConceptScore W4362588015C31972630 @default.
- W4362588015 hasConceptScore W4362588015C33923547 @default.
- W4362588015 hasConceptScore W4362588015C41008148 @default.
- W4362588015 hasConceptScore W4362588015C45347329 @default.
- W4362588015 hasConceptScore W4362588015C50644808 @default.
- W4362588015 hasConceptScore W4362588015C52622490 @default.
- W4362588015 hasConceptScore W4362588015C83546350 @default.
- W4362588015 hasIssue "4" @default.
- W4362588015 hasLocation W43625880151 @default.
- W4362588015 hasOpenAccess W4362588015 @default.
- W4362588015 hasPrimaryLocation W43625880151 @default.
- W4362588015 hasRelatedWork W1964120219 @default.
- W4362588015 hasRelatedWork W2000165426 @default.
- W4362588015 hasRelatedWork W2114557664 @default.
- W4362588015 hasRelatedWork W2144059113 @default.
- W4362588015 hasRelatedWork W2146076056 @default.
- W4362588015 hasRelatedWork W2385132419 @default.
- W4362588015 hasRelatedWork W2772780115 @default.
- W4362588015 hasRelatedWork W2811390910 @default.
- W4362588015 hasRelatedWork W2942471066 @default.