Feature Enhancement Based on CycleGAN for Nighttime Vehicle Detection

Shao, Xiaotao; Wei, Caike; Shen, Yan; Wang, Zhongli

doi:10.1109/access.2020.3046498

Cited by 39 publications

(18 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the visibility condition deteriorates, the detection performance of the aforementioned methods decreases because of the reduced visibility of object features such that specialized detection algorithms are required. For instance, for adverse weather conditions like snow or fog, Hassaballah et al, 2021 proposed a promising image enhancement strategy after which the aforementioned detectors can be applied, and, for nighttime, researchers studied whether a style transformation between nighttime and daylight images can be performed by a generative adversarial network (Shao et al, 2021;Lin et al, 2021). Even if the latter idea is promising, it is not mature enough to compete with the performances of specialized detection algorithms for nighttime (in terms of detection rates and computational efficiency).…”

Section: Vehicle Detectionmentioning

confidence: 99%

Provident vehicle detection at night for advanced driver assistance systems

Ewecker¹,

Asan

Ohnemus

et al. 2022

Auton Robot

View full text Add to dashboard Cite

In recent years, computer vision algorithms have become more powerful, which enabled technologies such as autonomous driving to evolve rapidly. However, current algorithms mainly share one limitation: They rely on directly visible objects. This is a significant drawback compared to human behavior, where visual cues caused by objects (e. g., shadows) are already used intuitively to retrieve information or anticipate occurring objects. While driving at night, this performance deficit becomes even more obvious: Humans already process the light artifacts caused by the headlamps of oncoming vehicles to estimate where they appear, whereas current object detection systems require that the oncoming vehicle is directly visible before it can be detected. Based on previous work on this subject, in this paper, we present a complete system that can detect light artifacts caused by the headlights of oncoming vehicles so that it detects that a vehicle is approaching providently (denoted as provident vehicle detection). For that, an entire algorithm architecture is investigated, including the detection in the image space, the three-dimensional localization, and the tracking of light artifacts. To demonstrate the usefulness of such an algorithm, the proposed algorithm is deployed in a test vehicle to use the detected light artifacts to control the glare-free high beam system proactively (react before the oncoming vehicle is directly visible). Using this experimental setting, the provident vehicle detection system’s time benefit compared to an in-production computer vision system is quantified. Additionally, the glare-free high beam use case provides a real-time and real-world visualization interface of the detection results by considering the adaptive headlamps as projectors. With this investigation of provident vehicle detection, we want to put awareness on the unconventional sensing task of detecting objects providently (detection based on observable visual cues the objects cause before they are visible) and further close the performance gap between human behavior and computer vision algorithms to bring autonomous and automated driving a step forward.

show abstract

Section: Vehicle Detectionmentioning

confidence: 99%

Provident vehicle detection at night for advanced driver assistance systems

Ewecker¹,

Asan

Ohnemus

et al. 2022

Auton Robot

View full text Add to dashboard Cite

show abstract

“…Mostly YOLO version 3 is best used for both pedestrian [10] and vehicle detection. By conducting the experiments [9], it is greatly shows the 'You Only Look Once' (YOLO) algorithm is well richer than the Faster R-CNN in terms accuracy and faster. Based on the Deep learning networks, YOLO is the end-to-end training are provided the great real time vehicle detector [8].…”

Section: Related To Workmentioning

confidence: 99%

Nighttime Vehicle Detection using Improved CycleGAN

Sathananthavathi¹,

Kandasamy²,

Rajamanickam³

2022

Preprint

View full text Add to dashboard Cite

Nighttime Vehicle Detection is the challenging task to be addressed properly in the Automatic Driving System. Detection of oncoming vehicles in time is essential for safe driving in the Automatic Driving System and this task is critical when the lighting condition is deprived. Detecting vehicle in nighttime is based on the high-level features by the automotive system. In order to address the problem of nighttime vehicle detection effectively, image translation by improved Cycle Generative Adversarial Network (GAN) architecture is proposed in this paper. The Cycle Generative Adversarial Network (CycleGAN) is the deep convolution neural network trained in the form of image-to-image translation tasks. Cycle GAN contains two GANs, first GAN is used for night to day translation and second GAN for day to night translation. The proposed improved Cycle GAN architecture is enriched with shallow Unet Generator in the first image translator GAN. Using this architecture, the input night time image is getting translated into day time image and hence supporting to extract high level of features for the vehicle detection module to detect the vehicles with high confidence probability. The proposed architecture is compared with the conventional cycle GAN architecture in terms of the vehicle detection accuracy. From the experimental results, it is observed that proposed improved Cycle GAN can able to detect vehicles in the perplex view condition.

show abstract

“…Since Goodfellow et al proposed the GAN [13], it has been applied in image processing successively, such as in singleimage super-resolution [1]- [3], image synthesis [4], [5], image restoration [6]- [10], object detection [11], etc. The GAN includes generators and discriminators.…”

Section: Related Work a Generative Adversarial Networkmentioning

confidence: 99%

Cross-View Image Translation Based on Local and Global Information Guidance

et al. 2021

Self Cite

View full text Add to dashboard Cite

The cross-view image translation task is aimed at generating scene images from arbitrary views. However, due to the great differences in the shapes and contents of the various views, the quality of the generated images is degraded. Small objects, such as vehicles' shapes and details, are not clear, which causes them to be structurally inconsistent with the semantic map used to guide the generation process. To solve this problem, we propose a novel generative adversarial network based on a local and global information processing module (LAGGAN) to recover the image's details and structures. The network will further combine the input viewpoint image and the target semantic segmentation map to guide the generation of the target image from another viewpoint. The proposed LAGGAN includes a two-stage generator and a parameter-sharing discriminator. LAGGAN uses a new local and global information processing module (LAG) to generate highquality images from various views. Moreover, we integrate dilated convolutions into the discriminator to capture the global context, which can enhance the discriminative ability and further adjust the LAG module. Therefore, most semantic information can be preserved, and the details of the target viewpoint images can be translated more sharply. Quantitative and qualitative evaluation on both CVUSA and Dayton datasets attest to the fact that our method, LAGGAN, presents satisfactory perceptual results and is comparable to state-of-the-art methods on the cross-view image translation task.INDEX TERMS Aerial images, cross-view image translation, generative adversarial networks (GANs), ground-level images, local and global information processing module.

show abstract

Feature Enhancement Based on CycleGAN for Nighttime Vehicle Detection

Cited by 39 publications

References 37 publications

Provident vehicle detection at night for advanced driver assistance systems

Provident vehicle detection at night for advanced driver assistance systems

Nighttime Vehicle Detection using Improved CycleGAN

Cross-View Image Translation Based on Local and Global Information Guidance

Contact Info

Product

Resources

About