ReSTiNet : An efficient deep learning approach to improve human detection accuracy
Human detection is an important task in computer vision. It is one of the most important tasks in global security and safety monitoring. In recent days, Deep Learning has improved human detection technology. Despite modern techniques, there are very few optimal techniques to construct networks with...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier B.V.
2023
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/38179/1/ReSTiNet_An%20efficient%20deep%20learning%20approach%20to%20improve%20human%20detection%20accuracy.pdf |
| _version_ | 1825815064302583808 |
|---|---|
| author | Sumit, Shahriar Shakir Dayang Rohaya, Awang Rambli Seyedali, Mirjalili Miah, Md Saef Ullah Muhammad Mudassir, Ejaz |
| author_facet | Sumit, Shahriar Shakir Dayang Rohaya, Awang Rambli Seyedali, Mirjalili Miah, Md Saef Ullah Muhammad Mudassir, Ejaz |
| author_sort | Sumit, Shahriar Shakir |
| collection | UMP |
| description | Human detection is an important task in computer vision. It is one of the most important tasks in global security and safety monitoring. In recent days, Deep Learning has improved human detection technology. Despite modern techniques, there are very few optimal techniques to construct networks with a small size, deep architecture, and fast training time while maintaining accuracy. ReSTiNet is a novel small convolutional neural network that overcomes the problems of network size, detection speed, and accuracy. The developed ReSTiNet contains fire modules by evaluating their number and position in the network to minimize the model parameters and network size. To improve the detection speed and accuracy of ReSTiNet, the residual block within the fire modules is carefully designed to increase the feature propagation and maximize the information flow in the network. The developed approach compresses the well-known Tiny-YOLO architecture while improving the following features: (i) small model size, (ii) faster detection speed, (iii) resolution of overfitting, and (iv) better performance than other compact networks such as SqueezeNet and MobileNet in terms of mAP on the Pascal VOC and MS COCO datasets. ReSTiNet is 10.7 MB, five times smaller than Tiny-YOLO. On Tesla k80, mAP is 27.3% for MS COCO and 63.74% for PASCAL VOC. The validation of the proposed ReSTiNet model has been done on INRIA person dataset using the Tesla K80. • All the necessary steps, algorithms, and mathematical formulas for building the net- work are provided. • The network is small in size but has a faster detection speed with high accuracy. |
| first_indexed | 2024-03-06T13:07:52Z |
| format | Article |
| id | UMPir38179 |
| institution | Universiti Malaysia Pahang |
| language | English |
| last_indexed | 2024-03-06T13:07:52Z |
| publishDate | 2023 |
| publisher | Elsevier B.V. |
| record_format | dspace |
| spelling | UMPir381792023-11-06T01:13:35Z http://umpir.ump.edu.my/id/eprint/38179/ ReSTiNet : An efficient deep learning approach to improve human detection accuracy Sumit, Shahriar Shakir Dayang Rohaya, Awang Rambli Seyedali, Mirjalili Miah, Md Saef Ullah Muhammad Mudassir, Ejaz QA75 Electronic computers. Computer science QA76 Computer software T Technology (General) TA Engineering (General). Civil engineering (General) Human detection is an important task in computer vision. It is one of the most important tasks in global security and safety monitoring. In recent days, Deep Learning has improved human detection technology. Despite modern techniques, there are very few optimal techniques to construct networks with a small size, deep architecture, and fast training time while maintaining accuracy. ReSTiNet is a novel small convolutional neural network that overcomes the problems of network size, detection speed, and accuracy. The developed ReSTiNet contains fire modules by evaluating their number and position in the network to minimize the model parameters and network size. To improve the detection speed and accuracy of ReSTiNet, the residual block within the fire modules is carefully designed to increase the feature propagation and maximize the information flow in the network. The developed approach compresses the well-known Tiny-YOLO architecture while improving the following features: (i) small model size, (ii) faster detection speed, (iii) resolution of overfitting, and (iv) better performance than other compact networks such as SqueezeNet and MobileNet in terms of mAP on the Pascal VOC and MS COCO datasets. ReSTiNet is 10.7 MB, five times smaller than Tiny-YOLO. On Tesla k80, mAP is 27.3% for MS COCO and 63.74% for PASCAL VOC. The validation of the proposed ReSTiNet model has been done on INRIA person dataset using the Tesla K80. • All the necessary steps, algorithms, and mathematical formulas for building the net- work are provided. • The network is small in size but has a faster detection speed with high accuracy. Elsevier B.V. 2023-01 Article PeerReviewed pdf en cc_by_4 http://umpir.ump.edu.my/id/eprint/38179/1/ReSTiNet_An%20efficient%20deep%20learning%20approach%20to%20improve%20human%20detection%20accuracy.pdf Sumit, Shahriar Shakir and Dayang Rohaya, Awang Rambli and Seyedali, Mirjalili and Miah, Md Saef Ullah and Muhammad Mudassir, Ejaz (2023) ReSTiNet : An efficient deep learning approach to improve human detection accuracy. MethodsX, 10 (101936). pp. 1-8. ISSN 2215-0161. (Published) https://doi.org/10.1016/j.mex.2022.101936 https://doi.org/10.1016/j.mex.2022.101936 |
| spellingShingle | QA75 Electronic computers. Computer science QA76 Computer software T Technology (General) TA Engineering (General). Civil engineering (General) Sumit, Shahriar Shakir Dayang Rohaya, Awang Rambli Seyedali, Mirjalili Miah, Md Saef Ullah Muhammad Mudassir, Ejaz ReSTiNet : An efficient deep learning approach to improve human detection accuracy |
| title | ReSTiNet : An efficient deep learning approach to improve human detection accuracy |
| title_full | ReSTiNet : An efficient deep learning approach to improve human detection accuracy |
| title_fullStr | ReSTiNet : An efficient deep learning approach to improve human detection accuracy |
| title_full_unstemmed | ReSTiNet : An efficient deep learning approach to improve human detection accuracy |
| title_short | ReSTiNet : An efficient deep learning approach to improve human detection accuracy |
| title_sort | restinet an efficient deep learning approach to improve human detection accuracy |
| topic | QA75 Electronic computers. Computer science QA76 Computer software T Technology (General) TA Engineering (General). Civil engineering (General) |
| url | http://umpir.ump.edu.my/id/eprint/38179/1/ReSTiNet_An%20efficient%20deep%20learning%20approach%20to%20improve%20human%20detection%20accuracy.pdf |
| work_keys_str_mv | AT sumitshahriarshakir restinetanefficientdeeplearningapproachtoimprovehumandetectionaccuracy AT dayangrohayaawangrambli restinetanefficientdeeplearningapproachtoimprovehumandetectionaccuracy AT seyedalimirjalili restinetanefficientdeeplearningapproachtoimprovehumandetectionaccuracy AT miahmdsaefullah restinetanefficientdeeplearningapproachtoimprovehumandetectionaccuracy AT muhammadmudassirejaz restinetanefficientdeeplearningapproachtoimprovehumandetectionaccuracy |