Matches in SemOpenAlex for { <https://semopenalex.org/work/W2783503594> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2783503594 endingPage "153" @default.
- W2783503594 startingPage "142" @default.
- W2783503594 abstract "Crowd distribution estimation has strong demands in surveillance applications, such as overcrowding detection, anomaly detection and traffic monitoring. Although a number of methods have been proposed for crowd counting, it is still a challenging task to estimate an accurate crowd distribution map which reflects the actual spatial intensity of the crowd in a real scene, due to the inhomogeneity of crowd distribution and the uncertainty of observation perspective. To address this problem, this paper proposes a multi-scale recursive convolutional neural network (MRCNN) based framework to map the image to its crowd distribution map. The proposed neural network is trained alternatively with two joint objectives, the estimation of crowd density map and perspective map. Since the scale size and scale variance of crowd are good cues for estimating both crowd density map and perspective map, formulating these two objectives together enables learning a strong feature representation for both tasks. By convolving a perspective-adaptive kernel on the crowd density map, we can generate a pixel-wise crowd distribution map in which the pixel value denotes the actual intensity of the crowd at the corresponding location in the real scene. An extension dataset from Shanghaitech crowd dataset B is introduced for the perspective map learning task, in which 700 images with about 3500 height-annotated pedestrians are labelled. Experimental results on Shanghaitech datasets (both A and B), UCF_CC_50 dataset and UCSD dataset demonstrate the effectiveness and reliability of our proposed approach." @default.
- W2783503594 created "2018-01-26" @default.
- W2783503594 creator A5002184499 @default.
- W2783503594 creator A5004299336 @default.
- W2783503594 creator A5025987638 @default.
- W2783503594 creator A5061817702 @default.
- W2783503594 date "2018-01-01" @default.
- W2783503594 modified "2023-09-29" @default.
- W2783503594 title "Crowd Distribution Estimation with Multi-scale Recursive Convolutional Neural Network" @default.
- W2783503594 cites W1910776219 @default.
- W2783503594 cites W2042322506 @default.
- W2783503594 cites W2049163084 @default.
- W2783503594 cites W2072232009 @default.
- W2783503594 cites W2096229530 @default.
- W2783503594 cites W2105206277 @default.
- W2783503594 cites W2116014710 @default.
- W2783503594 cites W2121468204 @default.
- W2783503594 cites W2123175289 @default.
- W2783503594 cites W2143043044 @default.
- W2783503594 cites W2155893237 @default.
- W2783503594 cites W2155916750 @default.
- W2783503594 cites W2158979073 @default.
- W2783503594 cites W2160624808 @default.
- W2783503594 cites W2161841955 @default.
- W2783503594 cites W2214802144 @default.
- W2783503594 cites W2463631526 @default.
- W2783503594 cites W2514654788 @default.
- W2783503594 cites W2519281173 @default.
- W2783503594 cites W2741077351 @default.
- W2783503594 cites W2962720716 @default.
- W2783503594 doi "https://doi.org/10.1007/978-3-319-73603-7_12" @default.
- W2783503594 hasPublicationYear "2018" @default.
- W2783503594 type Work @default.
- W2783503594 sameAs 2783503594 @default.
- W2783503594 citedByCount "3" @default.
- W2783503594 countsByYear W27835035942020 @default.
- W2783503594 countsByYear W27835035942022 @default.
- W2783503594 countsByYear W27835035942023 @default.
- W2783503594 crossrefType "book-chapter" @default.
- W2783503594 hasAuthorship W2783503594A5002184499 @default.
- W2783503594 hasAuthorship W2783503594A5004299336 @default.
- W2783503594 hasAuthorship W2783503594A5025987638 @default.
- W2783503594 hasAuthorship W2783503594A5061817702 @default.
- W2783503594 hasConcept C105795698 @default.
- W2783503594 hasConcept C119857082 @default.
- W2783503594 hasConcept C12713177 @default.
- W2783503594 hasConcept C138885662 @default.
- W2783503594 hasConcept C153180895 @default.
- W2783503594 hasConcept C154945302 @default.
- W2783503594 hasConcept C185429906 @default.
- W2783503594 hasConcept C205649164 @default.
- W2783503594 hasConcept C2776401178 @default.
- W2783503594 hasConcept C2778755073 @default.
- W2783503594 hasConcept C31972630 @default.
- W2783503594 hasConcept C33923547 @default.
- W2783503594 hasConcept C41008148 @default.
- W2783503594 hasConcept C41895202 @default.
- W2783503594 hasConcept C58640448 @default.
- W2783503594 hasConcept C71134354 @default.
- W2783503594 hasConcept C81363708 @default.
- W2783503594 hasConceptScore W2783503594C105795698 @default.
- W2783503594 hasConceptScore W2783503594C119857082 @default.
- W2783503594 hasConceptScore W2783503594C12713177 @default.
- W2783503594 hasConceptScore W2783503594C138885662 @default.
- W2783503594 hasConceptScore W2783503594C153180895 @default.
- W2783503594 hasConceptScore W2783503594C154945302 @default.
- W2783503594 hasConceptScore W2783503594C185429906 @default.
- W2783503594 hasConceptScore W2783503594C205649164 @default.
- W2783503594 hasConceptScore W2783503594C2776401178 @default.
- W2783503594 hasConceptScore W2783503594C2778755073 @default.
- W2783503594 hasConceptScore W2783503594C31972630 @default.
- W2783503594 hasConceptScore W2783503594C33923547 @default.
- W2783503594 hasConceptScore W2783503594C41008148 @default.
- W2783503594 hasConceptScore W2783503594C41895202 @default.
- W2783503594 hasConceptScore W2783503594C58640448 @default.
- W2783503594 hasConceptScore W2783503594C71134354 @default.
- W2783503594 hasConceptScore W2783503594C81363708 @default.
- W2783503594 hasLocation W27835035941 @default.
- W2783503594 hasOpenAccess W2783503594 @default.
- W2783503594 hasPrimaryLocation W27835035941 @default.
- W2783503594 hasRelatedWork W1504288058 @default.
- W2783503594 hasRelatedWork W2017205855 @default.
- W2783503594 hasRelatedWork W2048505601 @default.
- W2783503594 hasRelatedWork W2167293474 @default.
- W2783503594 hasRelatedWork W2331674254 @default.
- W2783503594 hasRelatedWork W2767651786 @default.
- W2783503594 hasRelatedWork W2979079341 @default.
- W2783503594 hasRelatedWork W3027997911 @default.
- W2783503594 hasRelatedWork W3042897387 @default.
- W2783503594 hasRelatedWork W4287776258 @default.
- W2783503594 isParatext "false" @default.
- W2783503594 isRetracted "false" @default.
- W2783503594 magId "2783503594" @default.
- W2783503594 workType "book-chapter" @default.