Street Object Detection from Synthesized and Processed Semantic Image: A Deep Learning Based Study

Bibliographic Details
Title: Street Object Detection from Synthesized and Processed Semantic Image: A Deep Learning Based Study
Authors: Parthaw Goswami, A. B. M. Aowlad Hossain
Source: Human-Centric Intelligent Systems, Vol 3, Iss 4, Pp 487-507 (2023)
Publisher Information: Springer Science and Business Media LLC, 2023.
Publication Year: 2023
Subject Terms: Conditional Generative Adversarial Network, Artificial intelligence, Object detection, Image Inpainting, 0211 other engineering and technologies, Convolutional neural network, Information technology, 02 engineering and technology, Pattern recognition (psychology), Pedestrian detection, Engineering, Object Detection, 11. Sustainability, 0202 electrical engineering, electronic engineering, information engineering, Image (mathematics), Image Enhancement Techniques, Street object detection, Image Recognition, Neural style transfer, Deep learning, Detector, QA75.5-76.95, Pedestrian, Transport engineering, 15. Life on land, T58.5-58.64, Computer science, Generative Adversarial Networks in Image Processing, 13. Climate action, Super-resolution, Electronic computers. Computer science, Computer Science, Physical Sciences, Telecommunications, Semantic image synthesis, Deep Learning in Computer Vision and Image Recognition, Image Synthesis, Texture Synthesis, Computer vision, Object (grammar), Computer Vision and Pattern Recognition
Description: Semantic image synthesis plays an important role in the development of Advanced Driver Assistance System (ADAS). Street objects detection might be erroneous during raining or when images from vehicle’s camera are blurred, which can cause serious accidents. Therefore, automatic and accurate street object detection is a demanding research scope. In this paper, a deep learning based framework is proposed and investigated for street object detection from synthesized and processed semantic image. Firstly, a Conditional Generative Adversarial Network (CGAN) has been used to create the realistic image. The brightness of the CGAN generated image has been increased using neural style transfer method. Furthermore, Enhanced Super-Resolution Generative Adversarial Networks (ESRGAN) based image enhancement concept has been used to improve the resolution of style-transferred images. These processed images exhibit better clarity and high fidelity which is impactful in the performance improvement of object detector. Finally, the synthesized and processed images were used as input in a Region-based Convolutional Neural Network (Faster R-CNN) and a MobileNet Single Shot Detector (MobileNetSSDv2) model separately for object detection. The widely used Cityscape dataset is used to investigate the performance of the proposed framework. The results analysis shows that the used synthesized and processed input improves the performance of the detectors than the unprocessed counterpart. A comparison of the proposed detection framework with related state of the art techniques is also found satisfactory with a mean average precision (mAP) around 32.6%, whereas most of the cases, mAPs are reported in the range of 20–28% for this particular dataset.
Document Type: Article
Other literature type
Language: English
ISSN: 2667-1336
DOI: 10.1007/s44230-023-00043-1
DOI: 10.60692/q9z84-zas84
DOI: 10.60692/ccb9m-hkv93
Access URL: https://doaj.org/article/43bf7faa22c4470c99709e5126970009
Rights: CC BY
Accession Number: edsair.doi.dedup.....9b904bb4fe7b00de9e6d9c6ad9b0125c
Database: OpenAIRE
Description
ISSN:26671336
DOI:10.1007/s44230-023-00043-1