Analysis of Object Detection and Visibility in Urban Environments with Unreal Engine 5
Analyse der Objektdetektion und Sichtbarkeit in städtischen Umgebungen mit Unreal Engine 5
Master thesis
The field of Adaptive Lighting Systems and Visual Processing develops innovative lighting assistance systems that aim to support the driver, particularly at night, and improve road safety. A central goal of this research is to enable adaptive lighting using camera systems and pixelated headlights, creating optimal visibility conditions for every object.
This master's thesis investigates how different urban scenes and objects influence object detection in traffic. A detection test will be developed for various objects in an urban environment. The goal is to determine the detection thresholds and the visibility of objects in these scenarios while considering the dynamics of pedestrians, such as inserting walking pedestrians into the scene. The test will be developed using Unreal Engine 5 (UE5) as a simulation platform to create a realistic, controllable, and repeatable test environment.
Work Packages:
1. Literature Review:
o Investigating relevant literature on object detection and brightness perception in urban environments.
o Analyzing existing models and their applicability to urban traffic situations.
2. Test Setup with Unreal Engine 5:
o Developing a detection test in an urban environment using Unreal Engine 5.
o Different urban scenes will be created in UE5, such as streets with varying lighting conditions and different objects like vehicles, pedestrians, and street signs.
o The virtual test environment will be designed so that various objects can appear in the scene and allow for quick switching between scenes and objects.
o Ensuring the light technical correctness of the scenes.
3. Data Collection:
o Collecting data on detection thresholds for different objects in various urban scenes. → Participant test.
4. Statistical Analysis:
o Analyzing the collected data to determine the visibility of objects in urban scenes.
o Applying statistical methods to examine the relationship between different urban scenes, objects, and detection probabilities.
5. Dynamic Objects:
o Investigating the dynamic component, such as the movement of pedestrians or vehicles in the urban scene, simulated in Unreal Engine 5.
o Analyzing how the movement of objects affects detection thresholds.
Research Questions:
1. How does the dynamics affect the required visibility level for different detection probabilities (70%, 90%, 95%) of objects?
2. How do different urban scenes and lighting conditions affect the required visibility level for the various detection probabilities (70%, 90%, 95%) of objects?
Please send me your application along with a brief exposé, in which you describe your approach to solving the topic and explain why you are particularly suited for this project. Include information on which relevant courses you have taken during your studies and which projects you have successfully completed that have prepared you for this topic. Your application should also include a cover letter.
Please send your documents by email to kunst@lichttechnik.tu-darmstadt.de .
