A dynamic discarding technique to increase speed and preserve accuracy for YOLOv3

Abstract

This paper proposes an acceleration technique to minimise the unnecessary operations on a state-of-the-art machine learning model and thus to improve the processing speed while maintaining the accuracy. After the study of the main bottlenecks that negatively affect the performance of convolutional neural networks, this paper designs and implements a discarding technique for YOLOv3-based algorithms to increase the speed and maintain accuracy. After applying the discarding technique, YOLOv3 can achieve a 22% of improvement in terms of speed. Moreover, the results of this new discarding technique were tested on Tiny-YOLOv3 with three output layers on an autonomous vehicle for pedestrian detection and it achieved an improvement of 48.7% in speed. The dynamic discarding technique just needs one training process to create the model and thus execute the approach, which preserves accuracy. The improved detector based on the discarding technique is able to readily alert the operator of the autonomous vehicle to take the emergency brake of the vehicle in order to avoid collision and consequently save lives.

Smartphone-based real-time object recognition architecture for portable and constrained systems

“…However, there is still an extra 26% to reduce to achieve 24 fps in a constrained environment. Several approaches could be taken to further reduce the inference time [19], such as quantisation techniques or faster feature Fig. 11 Comparison of our approach with a ideal model that achieves real-time recognition per frame extraction methods [10].…”

Hold surrounding’s key-you only look once version 7: a real-time pedestrian and vehicle detection algorithm in the low-signal-to-noise ratio infrared image

“…lack of an efficient feature fusion strategy and loss calculation method. At present, vehicle and pedestrian detection in the low-signal-to-noise ratio (SNR) infrared image still need to face the challenge of significant differences in object scale and appearance, 16 and further improvement is needed to improve the detection accuracy of the object detection algorithm.…”

“…However, when the object to be detected has interference factors, such as fine granularity, severe occlusion, and complex background, the YOLOv7 object detection algorithm is difficult to deal with effectively due to the lack of an efficient feature fusion strategy and loss calculation method. At present, vehicle and pedestrian detection in the low-signal-to-noise ratio (SNR) infrared image still need to face the challenge of significant differences in object scale and appearance, 16 and further improvement is needed to improve the detection accuracy of the object detection algorithm.…”

Machine-Learning-Based Carbon Dioxide Concentration Prediction for Hybrid Vehicles

“…In addition, the proposed UWS-LSTM model has been embedded and deployed in a real-world HV to properly demonstrate its practical application in a low-powered Internet of Things (IoT) device, providing an efficient, cost-effective solution without the need for using PEMS. As seen from previous studies, IoT devices are increasingly improving their capabilities, which, coupled with the optimisation of ML techniques, can drastically provoke a change in this paradigm from highly-capable workstations to low-powered devices [ 13 ]. The outcomes of this study have the potential to assist policy makers and vehicle manufacturers regarding CO 2 emission in HVs.…”