FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3164972083> ?p ?o ?g. }

• W3164972083 abstract "Significant progress on the crowd counting problem has been achieved by integrating larger context into convolutional neural networks (CNNs). This indicates that global scene context is essential, despite the seemingly bottom-up nature of the problem. This may be explained by the fact that context knowledge can adapt and improve local feature extraction to a given scene. In this paper, we therefore investigate the role of global context for crowd counting. Specifically, a pure transformer is used to extract features with global information from overlapping image patches. Inspired by classification, we add a context token to the input sequence, to facilitate information exchange with tokens corresponding to image patches throughout transformer layers. Due to the fact that transformers do not explicitly model the tried-and-true channel-wise interactions, we propose a token-attention module (TAM) to recalibrate encoded features through channel-wise attention informed by the context token. Beyond that, it is adopted to predict the total person count of the image through regression-token module (RTM). Extensive experiments demonstrate that our method achieves state-of-the-art performance on various datasets, including ShanghaiTech, UCF-QNRF, JHU-CROWD++ and NWPU. On the large-scale JHU-CROWD++ dataset, our method improves over the previous best results by 26.9% and 29.9% in terms of MAE and MSE, respectively." @default.
• W3164972083 created "2021-06-07" @default.
• W3164972083 creator A5001254143 @default.
• W3164972083 creator A5003065832 @default.
• W3164972083 creator A5010215983 @default.
• W3164972083 creator A5031578653 @default.
• W3164972083 creator A5031624673 @default.
• W3164972083 creator A5050696776 @default.
• W3164972083 date "2021-05-23" @default.
• W3164972083 modified "2023-09-24" @default.
• W3164972083 title "Boosting Crowd Counting with Transformers" @default.
• W3164972083 cites W1861492603 @default.
• W3164972083 cites W1910776219 @default.
• W3164972083 cites W1976959044 @default.
• W3164972083 cites W1978232622 @default.
• W3164972083 cites W1986388100 @default.
• W3164972083 cites W2037227137 @default.
• W3164972083 cites W2075875861 @default.
• W3164972083 cites W2101178587 @default.
• W3164972083 cites W2117539524 @default.
• W3164972083 cites W2120815373 @default.
• W3164972083 cites W2122243179 @default.
• W3164972083 cites W2151666244 @default.
• W3164972083 cites W2194775991 @default.
• W3164972083 cites W2463631526 @default.
• W3164972083 cites W2541389513 @default.
• W3164972083 cites W2548275288 @default.
• W3164972083 cites W2558630670 @default.
• W3164972083 cites W2729018917 @default.
• W3164972083 cites W2741077351 @default.
• W3164972083 cites W2769210209 @default.
• W3164972083 cites W2798489385 @default.
• W3164972083 cites W2798781811 @default.
• W3164972083 cites W2883929025 @default.
• W3164972083 cites W2884585870 @default.
• W3164972083 cites W2884960332 @default.
• W3164972083 cites W2886443245 @default.
• W3164972083 cites W2895051362 @default.
• W3164972083 cites W2914913933 @default.
• W3164972083 cites W2949333977 @default.
• W3164972083 cites W2952506564 @default.
• W3164972083 cites W2953433552 @default.
• W3164972083 cites W2962720716 @default.
• W3164972083 cites W2962742544 @default.
• W3164972083 cites W2962921175 @default.
• W3164972083 cites W2963341956 @default.
• W3164972083 cites W2963403868 @default.
• W3164972083 cites W2963420686 @default.
• W3164972083 cites W2963686699 @default.
• W3164972083 cites W2963693541 @default.
• W3164972083 cites W2964110616 @default.
• W3164972083 cites W2964209782 @default.
• W3164972083 cites W2964309882 @default.
• W3164972083 cites W2967069910 @default.
• W3164972083 cites W2967776630 @default.
• W3164972083 cites W2969620138 @default.
• W3164972083 cites W2970597249 @default.
• W3164972083 cites W2970971581 @default.
• W3164972083 cites W2975221988 @default.
• W3164972083 cites W2981436300 @default.
• W3164972083 cites W2982007926 @default.
• W3164972083 cites W2982014038 @default.
• W3164972083 cites W2982130202 @default.
• W3164972083 cites W2985775862 @default.
• W3164972083 cites W2987988567 @default.
• W3164972083 cites W2996703886 @default.
• W3164972083 cites W3004672782 @default.
• W3164972083 cites W3010021361 @default.
• W3164972083 cites W3027606690 @default.
• W3164972083 cites W3034785991 @default.
• W3164972083 cites W3035193053 @default.
• W3164972083 cites W3081099313 @default.
• W3164972083 cites W3082900845 @default.
• W3164972083 cites W3084883754 @default.
• W3164972083 cites W3092787802 @default.
• W3164972083 cites W3096609285 @default.
• W3164972083 cites W3097305524 @default.
• W3164972083 cites W3097407159 @default.
• W3164972083 cites W3097994591 @default.
• W3164972083 cites W3101165561 @default.
• W3164972083 cites W3106546328 @default.
• W3164972083 cites W3107554785 @default.
• W3164972083 cites W3107653507 @default.
• W3164972083 cites W3108829545 @default.
• W3164972083 cites W3109272290 @default.
• W3164972083 cites W3112728669 @default.
• W3164972083 cites W3113688308 @default.
• W3164972083 cites W3119786062 @default.
• W3164972083 cites W3119952960 @default.
• W3164972083 cites W3121523901 @default.
• W3164972083 cites W3130921824 @default.
• W3164972083 cites W3131500599 @default.
• W3164972083 cites W3137292936 @default.
• W3164972083 cites W3154723007 @default.
• W3164972083 cites W3170841864 @default.
• W3164972083 cites W3171516518 @default.
• W3164972083 cites W3176047859 @default.
• W3164972083 cites W3188511781 @default.
• W3164972083 doi "https://doi.org/10.48550/arxiv.2105.10926" @default.
• W3164972083 hasPublicationYear "2021" @default.

• next