Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288468366> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4288468366 endingPage "5924" @default.
- W4288468366 startingPage "5911" @default.
- W4288468366 abstract "Action recognition has become a current research hotspot in computer vision. Compared to other deep learning methods, Two-stream convolutional network structure achieves better performance in action recognition, which divides the network into spatial and temporal streams, using video frame images as well as dense optical streams in the network, respectively, to obtain the category labels. However, the two-stream network has some drawbacks, i.e., using dense optical flow as the input of the temporal stream, which is computationally expensive and extremely time-consuming for the current extraction algorithm and cannot meet the requirements of real-time tasks. In this paper, instead of the dense optical flow, the Motion Vectors (MVs) are used and extracted from the compressed domain as temporal features, which greatly reduces the extraction time. However, the motion pattern that MVs contain is coarser, which leads to low accuracy. In this paper, we propose two strategies to improve the accuracy: firstly, an accumulated strategy is used to enhance the motion information and continuity of MVs; secondly, knowledge distillation is used to fuse the spatial information into the temporal stream so that more information (e.g., motion details, colors, etc.) is obtainable. Experimental results show that the accuracy of MV can be greatly improved by the strategies proposed in this paper and the final recognition for human actions accuracy is guaranteed without using optical flow." @default.
- W4288468366 created "2022-07-29" @default.
- W4288468366 creator A5017535979 @default.
- W4288468366 creator A5020675859 @default.
- W4288468366 creator A5046881277 @default.
- W4288468366 creator A5066473046 @default.
- W4288468366 creator A5075528541 @default.
- W4288468366 creator A5088731382 @default.
- W4288468366 date "2022-01-01" @default.
- W4288468366 modified "2023-09-29" @default.
- W4288468366 title "Motion Enhanced Model Based on High-Level Spatial Features" @default.
- W4288468366 cites W3015029968 @default.
- W4288468366 cites W3022053426 @default.
- W4288468366 cites W3023769690 @default.
- W4288468366 cites W3124936226 @default.
- W4288468366 cites W3198461787 @default.
- W4288468366 cites W4246059507 @default.
- W4288468366 cites W4206073726 @default.
- W4288468366 doi "https://doi.org/10.32604/cmc.2022.031664" @default.
- W4288468366 hasPublicationYear "2022" @default.
- W4288468366 type Work @default.
- W4288468366 citedByCount "0" @default.
- W4288468366 crossrefType "journal-article" @default.
- W4288468366 hasAuthorship W4288468366A5017535979 @default.
- W4288468366 hasAuthorship W4288468366A5020675859 @default.
- W4288468366 hasAuthorship W4288468366A5046881277 @default.
- W4288468366 hasAuthorship W4288468366A5066473046 @default.
- W4288468366 hasAuthorship W4288468366A5075528541 @default.
- W4288468366 hasAuthorship W4288468366A5088731382 @default.
- W4288468366 hasBestOaLocation W42884683661 @default.
- W4288468366 hasConcept C104114177 @default.
- W4288468366 hasConcept C115961682 @default.
- W4288468366 hasConcept C119599485 @default.
- W4288468366 hasConcept C124101348 @default.
- W4288468366 hasConcept C127413603 @default.
- W4288468366 hasConcept C141353440 @default.
- W4288468366 hasConcept C153180895 @default.
- W4288468366 hasConcept C154945302 @default.
- W4288468366 hasConcept C155542232 @default.
- W4288468366 hasConcept C31972630 @default.
- W4288468366 hasConcept C41008148 @default.
- W4288468366 hasConcept C89198739 @default.
- W4288468366 hasConceptScore W4288468366C104114177 @default.
- W4288468366 hasConceptScore W4288468366C115961682 @default.
- W4288468366 hasConceptScore W4288468366C119599485 @default.
- W4288468366 hasConceptScore W4288468366C124101348 @default.
- W4288468366 hasConceptScore W4288468366C127413603 @default.
- W4288468366 hasConceptScore W4288468366C141353440 @default.
- W4288468366 hasConceptScore W4288468366C153180895 @default.
- W4288468366 hasConceptScore W4288468366C154945302 @default.
- W4288468366 hasConceptScore W4288468366C155542232 @default.
- W4288468366 hasConceptScore W4288468366C31972630 @default.
- W4288468366 hasConceptScore W4288468366C41008148 @default.
- W4288468366 hasConceptScore W4288468366C89198739 @default.
- W4288468366 hasIssue "3" @default.
- W4288468366 hasLocation W42884683661 @default.
- W4288468366 hasOpenAccess W4288468366 @default.
- W4288468366 hasPrimaryLocation W42884683661 @default.
- W4288468366 hasRelatedWork W2122456626 @default.
- W4288468366 hasRelatedWork W2123374136 @default.
- W4288468366 hasRelatedWork W2171116555 @default.
- W4288468366 hasRelatedWork W2511137960 @default.
- W4288468366 hasRelatedWork W2604231787 @default.
- W4288468366 hasRelatedWork W2687972263 @default.
- W4288468366 hasRelatedWork W2912737833 @default.
- W4288468366 hasRelatedWork W3084370450 @default.
- W4288468366 hasRelatedWork W3214088465 @default.
- W4288468366 hasRelatedWork W4281553171 @default.
- W4288468366 hasVolume "73" @default.
- W4288468366 isParatext "false" @default.
- W4288468366 isRetracted "false" @default.
- W4288468366 workType "article" @default.