Matches in SemOpenAlex for { <https://semopenalex.org/work/W3208596572> ?p ?o ?g. }
- W3208596572 endingPage "148448" @default.
- W3208596572 startingPage "148433" @default.
- W3208596572 abstract "Advances in moving object detection have been driven by the active application of deep learning methods. However, many existing models render superior detection accuracy at the cost of high computational complexity and slow inference speed. This fact has hindered the development of such models in mobile and embedded vision tasks, which need to be carried out in a timely fashion on a computationally limited platform. In this paper, we propose a super-fast (inference speed-154 fps) and lightweight (model size-1.45 MB) end-to-end 3D separable convolutional neural network with a multi-input multi-output (MIMO) strategy named “3DS_MM” for moving object detection. To improve detection accuracy, the proposed model adopts 3D convolution which is more suitable to extract both spatial and temporal information in video data than 2D convolution. To reduce model size and computational complexity, the standard 3D convolution is decomposed into depthwise and pointwise convolutions. Besides, we proposed a MIMO strategy to increase inference speed, which can take multiple frames as the network input and output multiple frames of detection results. Further, we conducted the scene dependent evaluation (SDE) and scene independent evaluation (SIE) on the benchmark CDnet2014 and DAVIS2016 datasets. Compared to stateof- the-art approaches, our proposed method significantly increases the inference speed, reduces the model size, meanwhile achieving the highest detection accuracy in the SDE setup and maintaining a competitive detection accuracy in the SIE setup." @default.
- W3208596572 created "2021-11-08" @default.
- W3208596572 creator A5018686979 @default.
- W3208596572 creator A5041229741 @default.
- W3208596572 creator A5051673693 @default.
- W3208596572 creator A5053401529 @default.
- W3208596572 creator A5060002817 @default.
- W3208596572 date "2021-01-01" @default.
- W3208596572 modified "2023-09-27" @default.
- W3208596572 title "A Fast Lightweight 3D Separable Convolutional Neural Network With Multi-Input Multi-Output for Moving Object Detection" @default.
- W3208596572 cites W1499877760 @default.
- W3208596572 cites W1522734439 @default.
- W3208596572 cites W1821655158 @default.
- W3208596572 cites W1994304612 @default.
- W3208596572 cites W1994634851 @default.
- W3208596572 cites W2040645214 @default.
- W3208596572 cites W2067813398 @default.
- W3208596572 cites W2069057506 @default.
- W3208596572 cites W2116076678 @default.
- W3208596572 cites W2127070222 @default.
- W3208596572 cites W2142412278 @default.
- W3208596572 cites W2277192409 @default.
- W3208596572 cites W2294874601 @default.
- W3208596572 cites W2417256080 @default.
- W3208596572 cites W2470139095 @default.
- W3208596572 cites W2480634901 @default.
- W3208596572 cites W2516881908 @default.
- W3208596572 cites W2525668722 @default.
- W3208596572 cites W2564998703 @default.
- W3208596572 cites W2606629906 @default.
- W3208596572 cites W2621275874 @default.
- W3208596572 cites W2624386319 @default.
- W3208596572 cites W2750515003 @default.
- W3208596572 cites W2751961297 @default.
- W3208596572 cites W2759692151 @default.
- W3208596572 cites W2766141718 @default.
- W3208596572 cites W2768086131 @default.
- W3208596572 cites W2774176625 @default.
- W3208596572 cites W2783946051 @default.
- W3208596572 cites W2806294667 @default.
- W3208596572 cites W2888845200 @default.
- W3208596572 cites W2889211735 @default.
- W3208596572 cites W2889347846 @default.
- W3208596572 cites W2891877542 @default.
- W3208596572 cites W2892141512 @default.
- W3208596572 cites W2892320136 @default.
- W3208596572 cites W2896633068 @default.
- W3208596572 cites W2896949148 @default.
- W3208596572 cites W2897606564 @default.
- W3208596572 cites W2901951655 @default.
- W3208596572 cites W2902239439 @default.
- W3208596572 cites W2902464468 @default.
- W3208596572 cites W2902993091 @default.
- W3208596572 cites W2903118147 @default.
- W3208596572 cites W2903164420 @default.
- W3208596572 cites W2909051837 @default.
- W3208596572 cites W2921997090 @default.
- W3208596572 cites W2925327970 @default.
- W3208596572 cites W2946472517 @default.
- W3208596572 cites W2947608396 @default.
- W3208596572 cites W2953452037 @default.
- W3208596572 cites W2957414648 @default.
- W3208596572 cites W2963131444 @default.
- W3208596572 cites W2963227409 @default.
- W3208596572 cites W2963529609 @default.
- W3208596572 cites W2963820951 @default.
- W3208596572 cites W2963868681 @default.
- W3208596572 cites W2966395992 @default.
- W3208596572 cites W2968179527 @default.
- W3208596572 cites W2983488121 @default.
- W3208596572 cites W2987878306 @default.
- W3208596572 cites W2996870660 @default.
- W3208596572 cites W3019468239 @default.
- W3208596572 cites W3035295069 @default.
- W3208596572 cites W3089901222 @default.
- W3208596572 cites W3090090415 @default.
- W3208596572 cites W3090739904 @default.
- W3208596572 cites W3093858573 @default.
- W3208596572 cites W3101795426 @default.
- W3208596572 cites W3105220622 @default.
- W3208596572 cites W3105536096 @default.
- W3208596572 cites W3112318124 @default.
- W3208596572 cites W3152141291 @default.
- W3208596572 cites W4231795354 @default.
- W3208596572 cites W4300179783 @default.
- W3208596572 doi "https://doi.org/10.1109/access.2021.3123975" @default.
- W3208596572 hasPublicationYear "2021" @default.
- W3208596572 type Work @default.
- W3208596572 sameAs 3208596572 @default.
- W3208596572 citedByCount "9" @default.
- W3208596572 countsByYear W32085965722022 @default.
- W3208596572 countsByYear W32085965722023 @default.
- W3208596572 crossrefType "journal-article" @default.
- W3208596572 hasAuthorship W3208596572A5018686979 @default.
- W3208596572 hasAuthorship W3208596572A5041229741 @default.
- W3208596572 hasAuthorship W3208596572A5051673693 @default.
- W3208596572 hasAuthorship W3208596572A5053401529 @default.
- W3208596572 hasAuthorship W3208596572A5060002817 @default.