Research Topic

Scene Understanding in Autonomous Driving

About this Research Topic

The last decade has witnessed increasing development of autonomous driving technologies through advancements in deep learning and artificial intelligence. The main purpose of autonomous driving is to improve driving safety and comfort by decreasing traffic accidents, which are primarily caused by human error. Autonomous driving enables a vehicle to operate autonomously by perceiving the traffic scene and predicting the movement of traffic actors. Accurate and efficient scene understanding, which is able to detect traffic participants, drivable areas, predict future events and plan future driving strategy, is a fundamental task of autonomous driving.

In autonomous driving, actionable information i.e. detecting and classifying objects (lanes, traffic lights, pedestrians, crossing lines, traffic signs, etc); determines the drivable space (distance to other cars, distance to pedestrians, etc) and allows for early predictions of upcoming events (if a car turns, if a car changes lane, if a pedestrian crosses the street, etc ) which means it plays an important role in autonomous driving. To achieve such actionable information, different modalities from various sources including Stereo, Lidar, Radar Ultrasonic, Flash, Event, and Thermal cameras together with various approaches such as supervised learning, reinforcement learning, etc. have been proposed. Furthermore, under the dynamic conditions of road traffic and practical implementation of on the edge devices, real-time inference, compact model, low energy systems with a high degree of accuracy are a critical requirement in autonomous driving.

This Research Topic welcomes contributions on topics including, but not limited to, the following:
• Driving Scene Understanding
• Objective Detection, Localization, and Tracking
• Efficient Video Feature Extraction
• Online Video Analysis
• Compressed/Light-Weight/Low-Cost Models
• Low Power/Energy Models
• Depth Estimation
• Point Cloud Reconstruction
• Scene Prediction
• Early Event Prediction
• Motion Planning, Navigation
• Models Fusing
• Databases, evaluation, and benchmarking in driving safety


Keywords: Scene Understanding, Scene Prediction, Driving Safety, Machine Learning, Deep Learning


Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.

The last decade has witnessed increasing development of autonomous driving technologies through advancements in deep learning and artificial intelligence. The main purpose of autonomous driving is to improve driving safety and comfort by decreasing traffic accidents, which are primarily caused by human error. Autonomous driving enables a vehicle to operate autonomously by perceiving the traffic scene and predicting the movement of traffic actors. Accurate and efficient scene understanding, which is able to detect traffic participants, drivable areas, predict future events and plan future driving strategy, is a fundamental task of autonomous driving.

In autonomous driving, actionable information i.e. detecting and classifying objects (lanes, traffic lights, pedestrians, crossing lines, traffic signs, etc); determines the drivable space (distance to other cars, distance to pedestrians, etc) and allows for early predictions of upcoming events (if a car turns, if a car changes lane, if a pedestrian crosses the street, etc ) which means it plays an important role in autonomous driving. To achieve such actionable information, different modalities from various sources including Stereo, Lidar, Radar Ultrasonic, Flash, Event, and Thermal cameras together with various approaches such as supervised learning, reinforcement learning, etc. have been proposed. Furthermore, under the dynamic conditions of road traffic and practical implementation of on the edge devices, real-time inference, compact model, low energy systems with a high degree of accuracy are a critical requirement in autonomous driving.

This Research Topic welcomes contributions on topics including, but not limited to, the following:
• Driving Scene Understanding
• Objective Detection, Localization, and Tracking
• Efficient Video Feature Extraction
• Online Video Analysis
• Compressed/Light-Weight/Low-Cost Models
• Low Power/Energy Models
• Depth Estimation
• Point Cloud Reconstruction
• Scene Prediction
• Early Event Prediction
• Motion Planning, Navigation
• Models Fusing
• Databases, evaluation, and benchmarking in driving safety


Keywords: Scene Understanding, Scene Prediction, Driving Safety, Machine Learning, Deep Learning


Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.

About Frontiers Research Topics

With their unique mixes of varied contributions from Original Research to Review Articles, Research Topics unify the most influential researchers, the latest key findings and historical advances in a hot research area! Find out more on how to host your own Frontiers Research Topic or contribute to one as an author.

Topic Editors

Loading..

Submission Deadlines

15 June 2021 Manuscript
15 July 2021 Manuscript Extension

Participating Journals

Manuscripts can be submitted to this Research Topic via the following journals:

Loading..

Topic Editors

Loading..

Submission Deadlines

15 June 2021 Manuscript
15 July 2021 Manuscript Extension

Participating Journals

Manuscripts can be submitted to this Research Topic via the following journals:

Loading..
Loading..

total views article views article downloads topic views

}
 
Top countries
Top referring sites
Loading..