Development of Fast Refinement Detectors on AI Edge Platforms

TL;DR

This paper develops real-time object detectors optimized for embedded AI edge platforms, balancing detection accuracy and inference speed using shallow CNN backbones and various architectures.

Contribution

It introduces a real-time object detection model tailored for embedded systems, utilizing shallow CNN backbones and intermediate layer optimization for balanced performance.

Findings

Achieved real-time detection on NVIDIA Drive PX2 and Jetson Xavier.

Demonstrated balanced accuracy and speed across multiple backbone architectures.

Provided publicly available code and models for embedded object detection.

Abstract

With the improvements in the object detection networks, several variations of object detection networks have been achieved impressive performance. However, the performance evaluation of most models has focused on detection accuracy, and performance verification is mostly based on high-end GPU hardware. In this paper, we propose real-time object detectors that guarantee balanced performance for real-time systems on embedded platforms. The proposed model utilizes the basic head structure of the RefineDet model, which is a variant of the single-shot object detector (SSD). In order to ensure real-time performance, CNN models with relatively shallow layers or fewer parameters have been used as the backbone structure. In addition to the basic VGGNet and ResNet structures, various backbone structures such as MobileNet, Xception, ResNeXt, Inception-SENet, and SE-ResNeXt have been used for this purpose. Successful training of object detection networks was achieved through an appropriate combination of intermediate layers. The accuracy of the proposed detector was estimated by the evaluation of the MS-COCO 2017 object detection dataset and the inference speed on the NVIDIA Drive PX2 and Jetson Xavier boards were tested to verify real-time performance in the embedded systems. The experiments show that the proposed models ensure balanced performance in terms of accuracy and inference speed in the embedded system environments. In addition, unlike the high-end GPUs, the use of embedded GPUs involves several additional concerns for efficient inference, which have been identified in this work. The codes and models are publicly available on the web (link).