Loading…
Video Semantic Segmentation Network with Low Latency Based on Deep Learning
Recently, new advances in deep learning algorithms have yielded some fascinating results in the field of computer vision technology. As a result, it can now perform activities that formerly required the use of human vision and the brain. Classification, object identification, and semantic segmentati...
Saved in:
Published in: | International journal of communication networks and information security 2023-12, Vol.15 (3), p.209-225 |
---|---|
Main Authors: | , |
Format: | Article |
Language: | English |
Subjects: | |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | |
---|---|
cites | |
container_end_page | 225 |
container_issue | 3 |
container_start_page | 209 |
container_title | International journal of communication networks and information security |
container_volume | 15 |
creator | Gowda D V, Channappa Kanagavalli, R. |
description | Recently, new advances in deep learning algorithms have yielded some fascinating results in the field of computer vision technology. As a result, it can now perform activities that formerly required the use of human vision and the brain. Classification, object identification, and semantic segmentation have all seen substantial advancements in deep learning architecture in the last few years. For still images and movies, there has been a major advancement in the field of semantic segmentation. In practical uses like autonomous vehicles, segmenting semantic video continues to be difficult due to high-performance standards, the high cost of convolutional neural networks (CNNs), and the significant need for low latency. An effective machine-learning environment will be developed to meet the performance and latency challenges outlined above. The use of deep learning architectures like SegNet and FlowNet2.0 on the CamVid dataset enables this environment to conduct pixel-wise semantic segmentation of video properties while maintaining low latency. As a result, it is ideally suited for real-world applications since it takes advantage of both SegNet and FlowNet topologies. The decision network determines whether an image frame should be processed by a segmentation network or an optical flow network based on the expected confidence score. In conjunction with adaptive scheduling of the key frame approach, this technique for decision-making can help to speed up the procedure. Using the ResNet50 SegNet model, a mean Intersection on Union (IoU) of "54.27 percent" and an average frame per second of "19.57" were observed. Aside from decision network and adaptive key frame sequencing, it was discovered that FlowNet2.0 increased the frames processed per second9(fps) to "30.19" on GPU with a mean IoU of "47.65%". Because the GPU was utilized "47.65%" of the time, this resulted. There has been an increase in the speed of the Video semantic segmentation network without sacrificing quality, as demonstrated by this improvement in performance. |
doi_str_mv | 10.17762/ijcnis.v15i3.6266 |
format | article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2918344814</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2918344814</sourcerecordid><originalsourceid>FETCH-LOGICAL-c141t-a8a89336a016e25b5277096e6bf7febc8c748f92b5a4b431324ee0a16b1561773</originalsourceid><addsrcrecordid>eNotkMlOwzAURS0EEqXwA6wssU7xFDtZQhlFBAsGsbMc96W4ULvYLlX_ntCwundx9J7uQeiUkglVSrJzt7DepckPLR2fSCblHhoxonghiXrf33VZkJqTQ3SU0oIQKQmpR-jhzc0g4GdYGp-d7ct8CT6b7ILHj5A3IX7ijcsfuAkb3JgM3m7xpUkwwz1xBbDCDZjonZ8fo4POfCU4-c8xer25fpneFc3T7f30oiksFTQXpjJVzbk0hEpgZVsypUgtQbad6qC1lVWi6mrWlka0glPOBAAxVLa0lP1aPkZnw91VDN9rSFkvwjr6_qVmNa24EBUVPcUGysaQUoROr6JbmrjVlOidND1I0ztp-k8a_wX0-WEM</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2918344814</pqid></control><display><type>article</type><title>Video Semantic Segmentation Network with Low Latency Based on Deep Learning</title><source>Freely Accessible Science Journals at publisher websites</source><creator>Gowda D V, Channappa ; Kanagavalli, R.</creator><creatorcontrib>Gowda D V, Channappa ; Kanagavalli, R.</creatorcontrib><description>Recently, new advances in deep learning algorithms have yielded some fascinating results in the field of computer vision technology. As a result, it can now perform activities that formerly required the use of human vision and the brain. Classification, object identification, and semantic segmentation have all seen substantial advancements in deep learning architecture in the last few years. For still images and movies, there has been a major advancement in the field of semantic segmentation. In practical uses like autonomous vehicles, segmenting semantic video continues to be difficult due to high-performance standards, the high cost of convolutional neural networks (CNNs), and the significant need for low latency. An effective machine-learning environment will be developed to meet the performance and latency challenges outlined above. The use of deep learning architectures like SegNet and FlowNet2.0 on the CamVid dataset enables this environment to conduct pixel-wise semantic segmentation of video properties while maintaining low latency. As a result, it is ideally suited for real-world applications since it takes advantage of both SegNet and FlowNet topologies. The decision network determines whether an image frame should be processed by a segmentation network or an optical flow network based on the expected confidence score. In conjunction with adaptive scheduling of the key frame approach, this technique for decision-making can help to speed up the procedure. Using the ResNet50 SegNet model, a mean Intersection on Union (IoU) of "54.27 percent" and an average frame per second of "19.57" were observed. Aside from decision network and adaptive key frame sequencing, it was discovered that FlowNet2.0 increased the frames processed per second9(fps) to "30.19" on GPU with a mean IoU of "47.65%". Because the GPU was utilized "47.65%" of the time, this resulted. There has been an increase in the speed of the Video semantic segmentation network without sacrificing quality, as demonstrated by this improvement in performance.</description><identifier>ISSN: 2076-0930</identifier><identifier>ISSN: 2073-607X</identifier><identifier>EISSN: 2073-607X</identifier><identifier>EISSN: 2076-0930</identifier><identifier>DOI: 10.17762/ijcnis.v15i3.6266</identifier><language>eng</language><publisher>Kohat: Kohat University of Science and Technology (KUST)</publisher><subject>Algorithms ; Artificial neural networks ; Brain research ; Classification ; Computer vision ; Datasets ; Decision making ; Deep learning ; Flow nets ; Frames per second ; Graphics processing units ; Machine learning ; Network latency ; Neural networks ; Optical flow (image analysis) ; Pixels ; Semantic segmentation ; Semantics</subject><ispartof>International journal of communication networks and information security, 2023-12, Vol.15 (3), p.209-225</ispartof><rights>Copyright Kohat University of Science and Technology (KUST) Sep 2023</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><creatorcontrib>Gowda D V, Channappa</creatorcontrib><creatorcontrib>Kanagavalli, R.</creatorcontrib><title>Video Semantic Segmentation Network with Low Latency Based on Deep Learning</title><title>International journal of communication networks and information security</title><description>Recently, new advances in deep learning algorithms have yielded some fascinating results in the field of computer vision technology. As a result, it can now perform activities that formerly required the use of human vision and the brain. Classification, object identification, and semantic segmentation have all seen substantial advancements in deep learning architecture in the last few years. For still images and movies, there has been a major advancement in the field of semantic segmentation. In practical uses like autonomous vehicles, segmenting semantic video continues to be difficult due to high-performance standards, the high cost of convolutional neural networks (CNNs), and the significant need for low latency. An effective machine-learning environment will be developed to meet the performance and latency challenges outlined above. The use of deep learning architectures like SegNet and FlowNet2.0 on the CamVid dataset enables this environment to conduct pixel-wise semantic segmentation of video properties while maintaining low latency. As a result, it is ideally suited for real-world applications since it takes advantage of both SegNet and FlowNet topologies. The decision network determines whether an image frame should be processed by a segmentation network or an optical flow network based on the expected confidence score. In conjunction with adaptive scheduling of the key frame approach, this technique for decision-making can help to speed up the procedure. Using the ResNet50 SegNet model, a mean Intersection on Union (IoU) of "54.27 percent" and an average frame per second of "19.57" were observed. Aside from decision network and adaptive key frame sequencing, it was discovered that FlowNet2.0 increased the frames processed per second9(fps) to "30.19" on GPU with a mean IoU of "47.65%". Because the GPU was utilized "47.65%" of the time, this resulted. There has been an increase in the speed of the Video semantic segmentation network without sacrificing quality, as demonstrated by this improvement in performance.</description><subject>Algorithms</subject><subject>Artificial neural networks</subject><subject>Brain research</subject><subject>Classification</subject><subject>Computer vision</subject><subject>Datasets</subject><subject>Decision making</subject><subject>Deep learning</subject><subject>Flow nets</subject><subject>Frames per second</subject><subject>Graphics processing units</subject><subject>Machine learning</subject><subject>Network latency</subject><subject>Neural networks</subject><subject>Optical flow (image analysis)</subject><subject>Pixels</subject><subject>Semantic segmentation</subject><subject>Semantics</subject><issn>2076-0930</issn><issn>2073-607X</issn><issn>2073-607X</issn><issn>2076-0930</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><recordid>eNotkMlOwzAURS0EEqXwA6wssU7xFDtZQhlFBAsGsbMc96W4ULvYLlX_ntCwundx9J7uQeiUkglVSrJzt7DepckPLR2fSCblHhoxonghiXrf33VZkJqTQ3SU0oIQKQmpR-jhzc0g4GdYGp-d7ct8CT6b7ILHj5A3IX7ijcsfuAkb3JgM3m7xpUkwwz1xBbDCDZjonZ8fo4POfCU4-c8xer25fpneFc3T7f30oiksFTQXpjJVzbk0hEpgZVsypUgtQbad6qC1lVWi6mrWlka0glPOBAAxVLa0lP1aPkZnw91VDN9rSFkvwjr6_qVmNa24EBUVPcUGysaQUoROr6JbmrjVlOidND1I0ztp-k8a_wX0-WEM</recordid><startdate>20231219</startdate><enddate>20231219</enddate><creator>Gowda D V, Channappa</creator><creator>Kanagavalli, R.</creator><general>Kohat University of Science and Technology (KUST)</general><scope>AAYXX</scope><scope>CITATION</scope><scope>3V.</scope><scope>4T-</scope><scope>4U-</scope><scope>7RQ</scope><scope>7XB</scope><scope>88F</scope><scope>88I</scope><scope>88K</scope><scope>8AL</scope><scope>8FE</scope><scope>8FG</scope><scope>8FK</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>GNUQQ</scope><scope>HCIFZ</scope><scope>JQ2</scope><scope>K7-</scope><scope>M0N</scope><scope>M1Q</scope><scope>M2P</scope><scope>M2T</scope><scope>P5Z</scope><scope>P62</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>Q9U</scope><scope>U9A</scope></search><sort><creationdate>20231219</creationdate><title>Video Semantic Segmentation Network with Low Latency Based on Deep Learning</title><author>Gowda D V, Channappa ; Kanagavalli, R.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c141t-a8a89336a016e25b5277096e6bf7febc8c748f92b5a4b431324ee0a16b1561773</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Algorithms</topic><topic>Artificial neural networks</topic><topic>Brain research</topic><topic>Classification</topic><topic>Computer vision</topic><topic>Datasets</topic><topic>Decision making</topic><topic>Deep learning</topic><topic>Flow nets</topic><topic>Frames per second</topic><topic>Graphics processing units</topic><topic>Machine learning</topic><topic>Network latency</topic><topic>Neural networks</topic><topic>Optical flow (image analysis)</topic><topic>Pixels</topic><topic>Semantic segmentation</topic><topic>Semantics</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Gowda D V, Channappa</creatorcontrib><creatorcontrib>Kanagavalli, R.</creatorcontrib><collection>CrossRef</collection><collection>ProQuest Central (Corporate)</collection><collection>Docstoc</collection><collection>University Readers</collection><collection>ProQuest Career & Technical Education Database</collection><collection>ProQuest Central (purchase pre-March 2016)</collection><collection>Military Database (Alumni Edition)</collection><collection>Science Database (Alumni Edition)</collection><collection>Telecommunications (Alumni Edition)</collection><collection>Computing Database (Alumni Edition)</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>ProQuest Central (Alumni) (purchase pre-March 2016)</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>Advanced Technologies & Aerospace Collection</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>ProQuest Central Student</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Computer Science Collection</collection><collection>Computer Science Database</collection><collection>Computing Database</collection><collection>Military Database</collection><collection>ProQuest Science Database</collection><collection>Telecommunications Database</collection><collection>Advanced Technologies & Aerospace Database</collection><collection>ProQuest Advanced Technologies & Aerospace Collection</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>ProQuest Central Basic</collection><jtitle>International journal of communication networks and information security</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Gowda D V, Channappa</au><au>Kanagavalli, R.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Video Semantic Segmentation Network with Low Latency Based on Deep Learning</atitle><jtitle>International journal of communication networks and information security</jtitle><date>2023-12-19</date><risdate>2023</risdate><volume>15</volume><issue>3</issue><spage>209</spage><epage>225</epage><pages>209-225</pages><issn>2076-0930</issn><issn>2073-607X</issn><eissn>2073-607X</eissn><eissn>2076-0930</eissn><abstract>Recently, new advances in deep learning algorithms have yielded some fascinating results in the field of computer vision technology. As a result, it can now perform activities that formerly required the use of human vision and the brain. Classification, object identification, and semantic segmentation have all seen substantial advancements in deep learning architecture in the last few years. For still images and movies, there has been a major advancement in the field of semantic segmentation. In practical uses like autonomous vehicles, segmenting semantic video continues to be difficult due to high-performance standards, the high cost of convolutional neural networks (CNNs), and the significant need for low latency. An effective machine-learning environment will be developed to meet the performance and latency challenges outlined above. The use of deep learning architectures like SegNet and FlowNet2.0 on the CamVid dataset enables this environment to conduct pixel-wise semantic segmentation of video properties while maintaining low latency. As a result, it is ideally suited for real-world applications since it takes advantage of both SegNet and FlowNet topologies. The decision network determines whether an image frame should be processed by a segmentation network or an optical flow network based on the expected confidence score. In conjunction with adaptive scheduling of the key frame approach, this technique for decision-making can help to speed up the procedure. Using the ResNet50 SegNet model, a mean Intersection on Union (IoU) of "54.27 percent" and an average frame per second of "19.57" were observed. Aside from decision network and adaptive key frame sequencing, it was discovered that FlowNet2.0 increased the frames processed per second9(fps) to "30.19" on GPU with a mean IoU of "47.65%". Because the GPU was utilized "47.65%" of the time, this resulted. There has been an increase in the speed of the Video semantic segmentation network without sacrificing quality, as demonstrated by this improvement in performance.</abstract><cop>Kohat</cop><pub>Kohat University of Science and Technology (KUST)</pub><doi>10.17762/ijcnis.v15i3.6266</doi><tpages>17</tpages></addata></record> |
fulltext | fulltext |
identifier | ISSN: 2076-0930 |
ispartof | International journal of communication networks and information security, 2023-12, Vol.15 (3), p.209-225 |
issn | 2076-0930 2073-607X 2073-607X 2076-0930 |
language | eng |
recordid | cdi_proquest_journals_2918344814 |
source | Freely Accessible Science Journals at publisher websites |
subjects | Algorithms Artificial neural networks Brain research Classification Computer vision Datasets Decision making Deep learning Flow nets Frames per second Graphics processing units Machine learning Network latency Neural networks Optical flow (image analysis) Pixels Semantic segmentation Semantics |
title | Video Semantic Segmentation Network with Low Latency Based on Deep Learning |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T03%3A09%3A10IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Video%20Semantic%20Segmentation%20Network%20with%20Low%20Latency%20Based%20on%20Deep%20Learning&rft.jtitle=International%20journal%20of%20communication%20networks%20and%20information%20security&rft.au=Gowda%20D%20V,%20Channappa&rft.date=2023-12-19&rft.volume=15&rft.issue=3&rft.spage=209&rft.epage=225&rft.pages=209-225&rft.issn=2076-0930&rft.eissn=2073-607X&rft_id=info:doi/10.17762/ijcnis.v15i3.6266&rft_dat=%3Cproquest_cross%3E2918344814%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c141t-a8a89336a016e25b5277096e6bf7febc8c748f92b5a4b431324ee0a16b1561773%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2918344814&rft_id=info:pmid/&rfr_iscdi=true |