Motoshi Kojima
Research Subjects
Research Interests
Motoshi Kojima has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
Daisuke Watanabe
Graduate School of Science, Technology and Innovation, Kobe University, Nada-Ku, Japan
1 shared publicationGo Matsukawa
Toyota Motor Corporation, Toyota, Japan
1 shared publication- Hiroko Tanabe
Graduate School of Humanities and Human Sciences, Hokkaido University
1 shared publication - Hiroshi Kawaguchi
Graduate School of Science, Technology and Innovation, Kobe University, Kobe, Japan
1 shared publication - Masahiko Yoshimoto
Kobe University
1 shared publication Masakazu Taichi
Graduate School of System Informatics, Kobe University, Nada-ku, Japan
1 shared publication
Latest Publications
- 3 December 2023
- Published by Springer Nature
- p. 240-253
Robotic-Human-Machine-Interface for Elderly Driving: Balancing Embodiment and Anthropomorphism for Improved Acceptance
- 3 December 2023
- Published by Springer Nature
- p. 240-253
Robotic-Human-Machine-Interface for Elderly Driving: Balancing Embodiment and Anthropomorphism for Improved Acceptance
- 13 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 14 (4) , 634-645
A 1.15-TOPS 6.57-TOPS/W Neural Network Processor for Multi-Scale Object Detection With Reduced Convolutional Operations
For automated driving cars, we present a 40-nm dedicated object detection processor with only three operations: 3 × 3 convolution, 1 × 1 convolution, and 4 × 4 deconvolution. Multi-scale object detection at high recognition accuracy is possible by virtue of the deconvolution feature and concatenation. The input memory for a feature map has 8-bit width. A multiplier for the inputs has 8-bit precision. Partial-sum memory, however, has 16-bit width to suppress detection accuracy deterioration in a layer with 1024 channels in the target network. By fixed-point bit precision, the external memory bandwidth and internal memory capacity are reduced. Optimized parallelization in input and output channels reduces the external memory bandwidth to 0.062 billion accesses per 1280 × 384 image with internal memory capacity of 400 kB. The detection error is 1.9% of that using single-precision floating point. The maximum operating frequency is 500 MHz at a supply voltage of 1 V. Its peak performance is 1.15 TOPS. The maximum energy efficiency is 6.57 TOPS/W at 174 MHz and 0.6 V.
- 13 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 14 (4) , 634-645
A 1.15-TOPS 6.57-TOPS/W Neural Network Processor for Multi-Scale Object Detection With Reduced Convolutional Operations
For automated driving cars, we present a 40-nm dedicated object detection processor with only three operations: 3 × 3 convolution, 1 × 1 convolution, and 4 × 4 deconvolution. Multi-scale object detection at high recognition accuracy is possible by virtue of the deconvolution feature and concatenation. The input memory for a feature map has 8-bit width. A multiplier for the inputs has 8-bit precision. Partial-sum memory, however, has 16-bit width to suppress detection accuracy deterioration in a layer with 1024 channels in the target network. By fixed-point bit precision, the external memory bandwidth and internal memory capacity are reduced. Optimized parallelization in input and output channels reduces the external memory bandwidth to 0.062 billion accesses per 1280 × 384 image with internal memory capacity of 400 kB. The detection error is 1.9% of that using single-precision floating point. The maximum operating frequency is 500 MHz at a supply voltage of 1 V. Its peak performance is 1.15 TOPS. The maximum energy efficiency is 6.57 TOPS/W at 174 MHz and 0.6 V.