Fig 2 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
Semantic segmentation, i.e. assigning each pixel in an image a class to which it belongs, can be a part of the implementation for the perception model of autonomous vehicles. Over the last years multiple powerful neural network architectures for solving this task have been developed. In this work, a simple lightweight but modular, fully convolution...
Context in source publication
Context 1
... all encoders behave somewhat similarly in accuracy, falling into a three percentage point window between 74.3% and 77.3% mIOU. Computational complexity varies wildly however, with the encoders mobilenet v2, resnet18 and resnet34 achieving real-time capability, i.e. inference faster than 20 fps. Fig. 2 shows the relationship between inference time and prediction accuracy. Modern networks are either very fast or very accurate, with the resnet34 based model falling somewhere in between. The older VGG classifiers fall short in terms of both accuracy as well as ...
Similar publications
![Fig.1. Autonomous vehicle architecture [9]](profile/Saad-Ul-Hassan-Syed/publication/359263639/figure/fig1/AS:1134250485121027@1647437797471/Autonomous-vehicle-architecture-9_Q320.jpg)
![Fig. 4. Point cloud estimation with lidar [4]](profile/Saad-Ul-Hassan-Syed/publication/359263639/figure/fig2/AS:1134250485129218@1647437797499/Point-cloud-estimation-with-lidar-4_Q320.jpg)
Lidar (Light detection and ranging) accounts for the key component for Autonomous driving architecture. It holds the backbone for current AVs (autonomous vehicles) because of two important characteristics object detection and Depth estimation. Due to its high sensitivity and accuracy, it has several applications in robotics, Advanced driver assista...
The preview of the road surface states is essential for improving the safety and the ride comfort of autonomous vehicles. The created dataset in this data article consists of 370151 road surface images captured under a wide range of road and weather conditions in China. The original pictures are acquired with a vehicle-mounted camera and then the p...
A tecnologia de Veículos Autónomos (AV) é anunciada como possível e segura no futuro. A implantação generalizada proporcionaria reduções drásticas nos acidentes e nas necessidades de estacionamento, duas vantagens para a sociedade, principalmente nas cidades. O papel do AV no total que é viajado e, portanto, no congestionamento, energia e emissões...
In this paper, the autonomous vehicle presented as a discrete-time Takagi-Sugeno fuzzy (T-S) model. We used the discrete-time T-S model since it is ready for the implementation unlike the continuous T-S fuzzy model. The main goal is to keep the autonomous vehicle in the centreline of the lane regardless the external disturbances. These disturbances...
While the majority of autonomous driving research has concentrated on everyday driving scenarios, further safety and performance improvements of autonomous vehicles require a focus on extreme driving conditions. In this context, autonomous racing is a new area of research that has been attracting considerable interest recently. Due to the fact that...
Citations
... The segmentation model architecturally is a scaled down variant of the Deeplab network [13] with a resnet34 backbone [14], specifically tuned to achieve real time capability on the research vehicle on an input resolution of 1280x384. With only 23.08M parameters, this model achieves an mIOU 1 of 76.4% on the Cityscapes [15] validation set with inference speed in excess of 20fps [16]. On the car, the camera images get passed through the model, thereby get segmented into the 19 classes of the Cityscapes training dataset and subsequently transformed into a top-down view using a simple linear transformation. ...
