DBGC: Dimension-Based Generic Convolution Block for Object Recognition
The object recognition concept is being widely used a result of increasing CCTV surveillance and the need for automatic object or activity detection from images or video. Increases in the use of various sensor networks have also raised the need of lightweight process frameworks. Much research has be...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/5/1780 |
| _version_ | 1827650700799115264 |
|---|---|
| author | Chirag Patel Dulari Bhatt Urvashi Sharma Radhika Patel Sharnil Pandya Kirit Modi Nagaraj Cholli Akash Patel Urvi Bhatt Muhammad Ahmed Khan Shubhankar Majumdar Mohd Zuhair Khushi Patel Syed Aziz Shah Hemant Ghayvat |
| author_facet | Chirag Patel Dulari Bhatt Urvashi Sharma Radhika Patel Sharnil Pandya Kirit Modi Nagaraj Cholli Akash Patel Urvi Bhatt Muhammad Ahmed Khan Shubhankar Majumdar Mohd Zuhair Khushi Patel Syed Aziz Shah Hemant Ghayvat |
| author_sort | Chirag Patel |
| collection | DOAJ |
| description | The object recognition concept is being widely used a result of increasing CCTV surveillance and the need for automatic object or activity detection from images or video. Increases in the use of various sensor networks have also raised the need of lightweight process frameworks. Much research has been carried out in this area, but the research scope is colossal as it deals with open-ended problems such as being able to achieve high accuracy in little time using lightweight process frameworks. Convolution Neural Networks and their variants are widely used in various computer vision activities, but most of the architectures of CNN are application-specific. There is always a need for generic architectures with better performance. This paper introduces the Dimension-Based Generic Convolution Block (DBGC), which can be used with any CNN to make the architecture generic and provide a dimension-wise selection of various height, width, and depth kernels. This single unit which uses the separable convolution concept provides multiple combinations using various dimension-based kernels. This single unit can be used for height-based, width-based, or depth-based dimensions; the same unit can even be used for height and width, width and depth, and depth and height dimensions. It can also be used for combinations involving all three dimensions of height, width, and depth. The main novelty of DBGC lies in the dimension selector block included in the proposed architecture. Proposed unoptimized kernel dimensions reduce FLOPs by around one third and also reduce the accuracy by around one half; semi-optimized kernel dimensions yield almost the same or higher accuracy with half the FLOPs of the original architecture, while optimized kernel dimensions provide 5 to 6% higher accuracy with around a 10 M reduction in FLOPs. |
| first_indexed | 2024-03-09T20:22:43Z |
| format | Article |
| id | doaj.art-d3e71a9dd1ff40aea7d769780e69a9dd |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T20:22:43Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-d3e71a9dd1ff40aea7d769780e69a9dd2023-11-23T23:46:07ZengMDPI AGSensors1424-82202022-02-01225178010.3390/s22051780DBGC: Dimension-Based Generic Convolution Block for Object RecognitionChirag Patel0Dulari Bhatt1Urvashi Sharma2Radhika Patel3Sharnil Pandya4Kirit Modi5Nagaraj Cholli6Akash Patel7Urvi Bhatt8Muhammad Ahmed Khan9Shubhankar Majumdar10Mohd Zuhair11Khushi Patel12Syed Aziz Shah13Hemant Ghayvat14Department of Computer Engineering, Devang Patel Institute of Advance Technology and Research (DEPSTAR), Faculty of Technology and Engineering (FTE), CHARUSAT Campus, Charotar University of Science and Technology (CHARUSAT), Changa 388421, IndiaParul University, Vadodara 382030, Gujarat, IndiaDepartment of Computer Engineering, Devang Patel Institute of Advance Technology and Research (DEPSTAR), Faculty of Technology and Engineering (FTE), CHARUSAT Campus, Charotar University of Science and Technology (CHARUSAT), Changa 388421, IndiaDepartment of Information Technology, Devang Patel Institute of Advance Technology and Research (DEPSTAR), Faculty of Technology and Engineering (FTE), CHARUSAT Campus, Charotar University of Science and Technology (CHARUSAT), Changa 388421, IndiaSymbiosis Institute of Technology, Symbiosis International (Deemed) University, Pune 412115, IndiaSankalchand Patel College of Engineering, Sankalchand Patel University, Visnagar 384315, IndiaDepartment of Information Science and Engineering, R. V. College of Engineering, Banglore 560059, IndiaDepartment of Information Technology, Devang Patel Institute of Advance Technology and Research (DEPSTAR), Faculty of Technology and Engineering (FTE), CHARUSAT Campus, Charotar University of Science and Technology (CHARUSAT), Changa 388421, IndiaDepartment of Information Technology, Devang Patel Institute of Advance Technology and Research (DEPSTAR), Faculty of Technology and Engineering (FTE), CHARUSAT Campus, Charotar University of Science and Technology (CHARUSAT), Changa 388421, IndiaDTU Health Tech Department of Health Technology, 247 99 Lyngby, DenmarkDepartment of Electronics and Communication Engineering, National Institute of Technology, Bijni Complex, Laitumkhrah, Shillong 793003, Meghalaya, IndiaDepartment of Computer Science and Engineering, Institute of Technology, Nirma University, Ahmedabad 382481, IndiaDepartment of Computer Engineering, Devang Patel Institute of Advance Technology and Research (DEPSTAR), Faculty of Technology and Engineering (FTE), CHARUSAT Campus, Charotar University of Science and Technology (CHARUSAT), Changa 388421, IndiaHealthcare Technology and Innovation Theme, Faculty Research Centre for Intelligent Healthcare, Coventry University, Richard Crossman Building, Coventry CV1 5RW, UKComputer Science Department, Faculty of Technology, Linnaeus University, P G Vejdes väg, 351 95 Växjö, SwedenThe object recognition concept is being widely used a result of increasing CCTV surveillance and the need for automatic object or activity detection from images or video. Increases in the use of various sensor networks have also raised the need of lightweight process frameworks. Much research has been carried out in this area, but the research scope is colossal as it deals with open-ended problems such as being able to achieve high accuracy in little time using lightweight process frameworks. Convolution Neural Networks and their variants are widely used in various computer vision activities, but most of the architectures of CNN are application-specific. There is always a need for generic architectures with better performance. This paper introduces the Dimension-Based Generic Convolution Block (DBGC), which can be used with any CNN to make the architecture generic and provide a dimension-wise selection of various height, width, and depth kernels. This single unit which uses the separable convolution concept provides multiple combinations using various dimension-based kernels. This single unit can be used for height-based, width-based, or depth-based dimensions; the same unit can even be used for height and width, width and depth, and depth and height dimensions. It can also be used for combinations involving all three dimensions of height, width, and depth. The main novelty of DBGC lies in the dimension selector block included in the proposed architecture. Proposed unoptimized kernel dimensions reduce FLOPs by around one third and also reduce the accuracy by around one half; semi-optimized kernel dimensions yield almost the same or higher accuracy with half the FLOPs of the original architecture, while optimized kernel dimensions provide 5 to 6% higher accuracy with around a 10 M reduction in FLOPs.https://www.mdpi.com/1424-8220/22/5/1780CNNseparable convolutionDBGCdimension-based kernels |
| spellingShingle | Chirag Patel Dulari Bhatt Urvashi Sharma Radhika Patel Sharnil Pandya Kirit Modi Nagaraj Cholli Akash Patel Urvi Bhatt Muhammad Ahmed Khan Shubhankar Majumdar Mohd Zuhair Khushi Patel Syed Aziz Shah Hemant Ghayvat DBGC: Dimension-Based Generic Convolution Block for Object Recognition Sensors CNN separable convolution DBGC dimension-based kernels |
| title | DBGC: Dimension-Based Generic Convolution Block for Object Recognition |
| title_full | DBGC: Dimension-Based Generic Convolution Block for Object Recognition |
| title_fullStr | DBGC: Dimension-Based Generic Convolution Block for Object Recognition |
| title_full_unstemmed | DBGC: Dimension-Based Generic Convolution Block for Object Recognition |
| title_short | DBGC: Dimension-Based Generic Convolution Block for Object Recognition |
| title_sort | dbgc dimension based generic convolution block for object recognition |
| topic | CNN separable convolution DBGC dimension-based kernels |
| url | https://www.mdpi.com/1424-8220/22/5/1780 |
| work_keys_str_mv | AT chiragpatel dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT dularibhatt dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT urvashisharma dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT radhikapatel dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT sharnilpandya dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT kiritmodi dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT nagarajcholli dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT akashpatel dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT urvibhatt dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT muhammadahmedkhan dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT shubhankarmajumdar dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT mohdzuhair dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT khushipatel dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT syedazizshah dbgcdimensionbasedgenericconvolutionblockforobjectrecognition AT hemantghayvat dbgcdimensionbasedgenericconvolutionblockforobjectrecognition |