Filters








6 Hits in 1.6 sec

A Real-Time FPGA Accelerator Based on Winograd Algorithm for Underwater Object Detection

Liangwei Cai, Ceng Wang, Yuan Xu
2021 Electronics  
Real-time object detection is a challenging but crucial task for autonomous underwater vehicles because of the complex underwater imaging environment.  ...  To overcome the difficulties caused by these problems to underwater object detection, an end-to-end CNN network combined U-Net and MobileNetV3-SSDLite is proposed.  ...  The proposed underwater object detection CNN network combined U-Net and MobileNetV3-SSDLite is successfully implemented into the FPGA of our underwater robot, where it can achieve real-time object detection  ... 
doi:10.3390/electronics10232889 fatcat:7p6woc4g6jc6tozls7ivi5h2py

Layer-specific Optimization for Mixed Data Flow with Mixed Precision in FPGA Design for CNN-based Object Detectors

Duy Thanh Nguyen, Hyun Kim, Hyuk-Jae Lee
2020 IEEE transactions on circuits and systems for video technology (Print)  
The model size is reduced by 22.66-28.93 times compared to that in a full-precision network with a negligible degradation of accuracy on VOC, COCO, and ImageNet datasets.  ...  The mixed data flow aims to minimize the off-chip access while demanding a minimal on-chip memory (BRAM) resource of an FPGA device.  ...  For achieving real-time operation, numerous FPGA designs are available for a YOLO CNN [7] - [11] . The previous designs in [7] , [10] , and [11] achieve a real-time throughput.  ... 
doi:10.1109/tcsvt.2020.3020569 fatcat:wdypzy6bufhuzj7vcs7arx5wem

2021 Index IEEE Transactions on Very Large Scale Integration (VLSI) Systems Vol. 29

2021 IEEE Transactions on Very Large Scale Integration (vlsi) Systems  
Lyu, F., +, TVLSI July 2021 1470-1474 Real-Time SSDLite Object Detection on FPGA.  ...  ., +, TVLSI Feb. 2021 333-346 IMCA: An Efficient In-Memory Convolution Accelerator. Yantir, H.E., +, TVLSI March 2021 447-460 Real-Time SSDLite Object Detection on FPGA.  ... 
doi:10.1109/tvlsi.2021.3136367 fatcat:fwqswbyzejgfhgbzywrvsf2qgi

On-Device Object Detection for More Efficient and Privacy-Compliant Visual Perception in Context-Aware Systems

Ivan Rodriguez-Conde, Celso Campos, Florentino Fdez-Riverola
2021 Applied Sciences  
providing users with a smoother and better-tailored experience, with no need of sharing their data with an outsourced service.  ...  Framed in that novel paradigm, this work presents a review of the recent advances made along those lines in object detection, providing a comprehensive study of the most relevant lightweight CNN-based  ...  Healthcare - - [56] 2019 Vehicle and pedestrian detection Generic - • Real-time execution • High accuracy [57] 2019 Object detection in UAV imagery Generic - • Real-time execution  ... 
doi:10.3390/app11199173 fatcat:cncucjelmrgv3mdortgxax2qly

Cracking open the DNN black-box

John Emmons, Sadjad Fouladi, Ganesh Ananthanarayanan, Shivaram Venkataraman, Silvio Savarese, Keith Winstein
2019 Proceedings of the 2019 Workshop on Hot Topics in Video Analytics and Intelligent Edges - HotEdgeVideo'19  
Despite rapid advances in system design, existing systems treat DNNs largely as "black boxes" and either deploy models entirely on a camera or compress videos for analysis in the cloud.  ...  We present promising results from preliminary work in efficiently encoding the intermediate activations sent between layers of a neural network and describe opportunities for further research.  ...  Modern DNNs, however, require special purpose hardware accelerators (e.g. GPUs, FPGAs [23], TPUs [24] ) to run in real time (≥ 30 frames/sec) on HD videos (≥ 720p resolution).  ... 
doi:10.1145/3349614.3356023 dblp:conf/mobicom/EmmonsFAVSW19 fatcat:z4psoigzdra4jlqtgskx5xbhga

FPGA Based Embedded Neural Network Object Detector

Lukas Baischer, Axel Jantsch
2021
Since FPGAs provide high flexibility in combination with low power consumption, FPGAs are a promising candidate for implementing a real-time capable object detector for edge devices.Therefore, this work  ...  However, the enormous computational effort that CPUs cannot handle in real-time is a crucial disadvantage of object detection based on deep learning compared to conventional algorithms.  ...  the given real-time object detection task.  ... 
doi:10.34726/hss.2021.69314 fatcat:becnaaulgze4pf7hmkba7lmfni