Compact Spatial Pyramid Pooling Deep Convolutional Neural Network Based Hand Gestures Decoder
Current deep learning convolutional neural network (DCNN) -based hand gesture detectors with acute precision demand incredibly high-performance computing power. Although DCNN-based detectors are capable of accurate classification, the sheer computing power needed for this form of classification make...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-11-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/21/7898 |
| _version_ | 1797548548277927936 |
|---|---|
| author | Akm Ashiquzzaman Hyunmin Lee Kwangki Kim Hye-Young Kim Jaehyung Park Jinsul Kim |
| author_facet | Akm Ashiquzzaman Hyunmin Lee Kwangki Kim Hye-Young Kim Jaehyung Park Jinsul Kim |
| author_sort | Akm Ashiquzzaman |
| collection | DOAJ |
| description | Current deep learning convolutional neural network (DCNN) -based hand gesture detectors with acute precision demand incredibly high-performance computing power. Although DCNN-based detectors are capable of accurate classification, the sheer computing power needed for this form of classification makes it very difficult to run with lower computational power in remote environments. Moreover, classical DCNN architectures have a fixed number of input dimensions, which forces preprocessing, thus making it impractical for real-world applications. In this research, a practical DCNN with an optimized architecture is proposed with DCNN filter/node pruning, and spatial pyramid pooling (SPP) is introduced in order to make the model input dimension-invariant. This compact SPP-DCNN module uses <inline-formula><math display="inline"><semantics><mrow><mn>65</mn><mo>%</mo></mrow></semantics></math></inline-formula> fewer parameters than traditional classifiers and operates almost <inline-formula><math display="inline"><semantics><mrow><mn>3</mn><mo>×</mo></mrow></semantics></math></inline-formula> faster than classical models. Moreover, the new improved proposed algorithm, which decodes gestures or sign language finger-spelling from videos, gave a benchmark highest accuracy with the fastest processing speed. This proposed method paves the way for various practical and applied hand gesture input-based human-computer interaction (HCI) applications. |
| first_indexed | 2024-03-10T15:01:02Z |
| format | Article |
| id | doaj.art-6c193d1151f04a389c94f70833d60204 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T15:01:02Z |
| publishDate | 2020-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-6c193d1151f04a389c94f70833d602042023-11-20T20:08:50ZengMDPI AGApplied Sciences2076-34172020-11-011021789810.3390/app10217898Compact Spatial Pyramid Pooling Deep Convolutional Neural Network Based Hand Gestures DecoderAkm Ashiquzzaman0Hyunmin Lee1Kwangki Kim2Hye-Young Kim3Jaehyung Park4Jinsul Kim5Department of ICT Convergence System Engineering, Chonnam National University, Gwangju 61186, KoreaHuman IT Convergence Research Center, Korea Electronics Technology Institute, Gyeonggi-do 13509, KoreaSchool of IT Convergence, Korea Nazarene University, Chungcheongnam-do 31172, KoreaSchool of Game/Game Software, Hongik University, Seoul 04066, KoreaDepartment of ICT Convergence System Engineering, Chonnam National University, Gwangju 61186, KoreaDepartment of ICT Convergence System Engineering, Chonnam National University, Gwangju 61186, KoreaCurrent deep learning convolutional neural network (DCNN) -based hand gesture detectors with acute precision demand incredibly high-performance computing power. Although DCNN-based detectors are capable of accurate classification, the sheer computing power needed for this form of classification makes it very difficult to run with lower computational power in remote environments. Moreover, classical DCNN architectures have a fixed number of input dimensions, which forces preprocessing, thus making it impractical for real-world applications. In this research, a practical DCNN with an optimized architecture is proposed with DCNN filter/node pruning, and spatial pyramid pooling (SPP) is introduced in order to make the model input dimension-invariant. This compact SPP-DCNN module uses <inline-formula><math display="inline"><semantics><mrow><mn>65</mn><mo>%</mo></mrow></semantics></math></inline-formula> fewer parameters than traditional classifiers and operates almost <inline-formula><math display="inline"><semantics><mrow><mn>3</mn><mo>×</mo></mrow></semantics></math></inline-formula> faster than classical models. Moreover, the new improved proposed algorithm, which decodes gestures or sign language finger-spelling from videos, gave a benchmark highest accuracy with the fastest processing speed. This proposed method paves the way for various practical and applied hand gesture input-based human-computer interaction (HCI) applications.https://www.mdpi.com/2076-3417/10/21/7898deep learningconvolutional neural networkhand gesture recognitionneural network pruningoptimization |
| spellingShingle | Akm Ashiquzzaman Hyunmin Lee Kwangki Kim Hye-Young Kim Jaehyung Park Jinsul Kim Compact Spatial Pyramid Pooling Deep Convolutional Neural Network Based Hand Gestures Decoder Applied Sciences deep learning convolutional neural network hand gesture recognition neural network pruning optimization |
| title | Compact Spatial Pyramid Pooling Deep Convolutional Neural Network Based Hand Gestures Decoder |
| title_full | Compact Spatial Pyramid Pooling Deep Convolutional Neural Network Based Hand Gestures Decoder |
| title_fullStr | Compact Spatial Pyramid Pooling Deep Convolutional Neural Network Based Hand Gestures Decoder |
| title_full_unstemmed | Compact Spatial Pyramid Pooling Deep Convolutional Neural Network Based Hand Gestures Decoder |
| title_short | Compact Spatial Pyramid Pooling Deep Convolutional Neural Network Based Hand Gestures Decoder |
| title_sort | compact spatial pyramid pooling deep convolutional neural network based hand gestures decoder |
| topic | deep learning convolutional neural network hand gesture recognition neural network pruning optimization |
| url | https://www.mdpi.com/2076-3417/10/21/7898 |
| work_keys_str_mv | AT akmashiquzzaman compactspatialpyramidpoolingdeepconvolutionalneuralnetworkbasedhandgesturesdecoder AT hyunminlee compactspatialpyramidpoolingdeepconvolutionalneuralnetworkbasedhandgesturesdecoder AT kwangkikim compactspatialpyramidpoolingdeepconvolutionalneuralnetworkbasedhandgesturesdecoder AT hyeyoungkim compactspatialpyramidpoolingdeepconvolutionalneuralnetworkbasedhandgesturesdecoder AT jaehyungpark compactspatialpyramidpoolingdeepconvolutionalneuralnetworkbasedhandgesturesdecoder AT jinsulkim compactspatialpyramidpoolingdeepconvolutionalneuralnetworkbasedhandgesturesdecoder |