A Full Featured Configurable Accelerator for Object Detection With YOLO
Object detection and classification is an essential task of computer vision. A very efficient algorithm for detection and classification is YOLO (You Look Only Once). We consider hardware architectures to run YOLO in real-time on embedded platforms. Designing a new dedicated accelerator for each new...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9435338/ |
| _version_ | 1818671343595421696 |
|---|---|
| author | Daniel Pestana Pedro R. Miranda Joao D. Lopes Rui P. Duarte Mario P. Vestias Horacio C. Neto Jose T. De Sousa |
| author_facet | Daniel Pestana Pedro R. Miranda Joao D. Lopes Rui P. Duarte Mario P. Vestias Horacio C. Neto Jose T. De Sousa |
| author_sort | Daniel Pestana |
| collection | DOAJ |
| description | Object detection and classification is an essential task of computer vision. A very efficient algorithm for detection and classification is YOLO (You Look Only Once). We consider hardware architectures to run YOLO in real-time on embedded platforms. Designing a new dedicated accelerator for each new version of YOLO is not feasible given the fast delivery of new versions. This work’s primary goal is to design a configurable and scalable core for creating specific object detection and classification systems based on YOLO, targeting embedded platforms. The core accelerates the execution of all the algorithm steps, including pre-processing, model inference and post-processing. It considers a fixed-point format, linearised activation functions, batch-normalisation, folding, and a hardware structure that exploits most of the available parallelism in CNN processing. The proposed core is configured for real-time execution of YOLOv3-Tiny and YOLOv4-Tiny, integrated into a RISC-V-based system-on-chip architecture and prototyped in an UltraScale XCKU040 FPGA (Field Programmable Gate Array). The solution achieves a performance of 32 and 31 frames per second for YOLOv3-Tiny and YOLOv4-Tiny, respectively, with a 16-bit fixed-point format. Compared to previous proposals, it improves the frame rate at a higher performance efficiency. The performance, area efficiency and configurability of the proposed core enable the fast development of real-time YOLO-based object detectors on embedded systems. |
| first_indexed | 2024-12-17T07:22:30Z |
| format | Article |
| id | doaj.art-6769f844dca2477b83480a2b03ef65a3 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-17T07:22:30Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-6769f844dca2477b83480a2b03ef65a32022-12-21T21:58:43ZengIEEEIEEE Access2169-35362021-01-019758647587710.1109/ACCESS.2021.30818189435338A Full Featured Configurable Accelerator for Object Detection With YOLODaniel Pestana0Pedro R. Miranda1Joao D. Lopes2https://orcid.org/0000-0002-8903-9715Rui P. Duarte3https://orcid.org/0000-0002-7060-4745Mario P. Vestias4https://orcid.org/0000-0001-8556-4507Horacio C. Neto5https://orcid.org/0000-0002-3621-8322Jose T. De Sousa6https://orcid.org/0000-0001-7525-7546INESC-ID, Instituto Superior Técnico, Universidade de Lisboa, Lisboa, PortugalINESC-ID, Instituto Superior Técnico, Universidade de Lisboa, Lisboa, PortugalINESC-ID, Instituto Superior Técnico, Universidade de Lisboa, Lisboa, PortugalINESC-ID, Instituto Superior Técnico, Universidade de Lisboa, Lisboa, PortugalINESC-ID, Instituto Superior de Engenharia de Lisboa, Instituto Politécnico de Lisboa, Lisboa, PortugalINESC-ID, Instituto Superior Técnico, Universidade de Lisboa, Lisboa, PortugalINESC-ID, Instituto Superior Técnico, Universidade de Lisboa, Lisboa, PortugalObject detection and classification is an essential task of computer vision. A very efficient algorithm for detection and classification is YOLO (You Look Only Once). We consider hardware architectures to run YOLO in real-time on embedded platforms. Designing a new dedicated accelerator for each new version of YOLO is not feasible given the fast delivery of new versions. This work’s primary goal is to design a configurable and scalable core for creating specific object detection and classification systems based on YOLO, targeting embedded platforms. The core accelerates the execution of all the algorithm steps, including pre-processing, model inference and post-processing. It considers a fixed-point format, linearised activation functions, batch-normalisation, folding, and a hardware structure that exploits most of the available parallelism in CNN processing. The proposed core is configured for real-time execution of YOLOv3-Tiny and YOLOv4-Tiny, integrated into a RISC-V-based system-on-chip architecture and prototyped in an UltraScale XCKU040 FPGA (Field Programmable Gate Array). The solution achieves a performance of 32 and 31 frames per second for YOLOv3-Tiny and YOLOv4-Tiny, respectively, with a 16-bit fixed-point format. Compared to previous proposals, it improves the frame rate at a higher performance efficiency. The performance, area efficiency and configurability of the proposed core enable the fast development of real-time YOLO-based object detectors on embedded systems.https://ieeexplore.ieee.org/document/9435338/Object detectionconvolutional neural networkFPGAlightweight YOLO |
| spellingShingle | Daniel Pestana Pedro R. Miranda Joao D. Lopes Rui P. Duarte Mario P. Vestias Horacio C. Neto Jose T. De Sousa A Full Featured Configurable Accelerator for Object Detection With YOLO IEEE Access Object detection convolutional neural network FPGA lightweight YOLO |
| title | A Full Featured Configurable Accelerator for Object Detection With YOLO |
| title_full | A Full Featured Configurable Accelerator for Object Detection With YOLO |
| title_fullStr | A Full Featured Configurable Accelerator for Object Detection With YOLO |
| title_full_unstemmed | A Full Featured Configurable Accelerator for Object Detection With YOLO |
| title_short | A Full Featured Configurable Accelerator for Object Detection With YOLO |
| title_sort | full featured configurable accelerator for object detection with yolo |
| topic | Object detection convolutional neural network FPGA lightweight YOLO |
| url | https://ieeexplore.ieee.org/document/9435338/ |
| work_keys_str_mv | AT danielpestana afullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT pedrormiranda afullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT joaodlopes afullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT ruipduarte afullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT mariopvestias afullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT horaciocneto afullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT josetdesousa afullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT danielpestana fullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT pedrormiranda fullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT joaodlopes fullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT ruipduarte fullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT mariopvestias fullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT horaciocneto fullfeaturedconfigurableacceleratorforobjectdetectionwithyolo AT josetdesousa fullfeaturedconfigurableacceleratorforobjectdetectionwithyolo |