Deep Learning + Software Engineering: Effect of noisy map labels on the map perception task

Research topic/area

Deep Learning in Autonomous Driving

Type of thesis

Bachelor / Master

Start time

-

Application deadline

01.08.2025

Duration of the thesis

3 - 8 Monate

Description

Current state-of-the-art map construction methods such as MapTRv2 use sensor data (360° surround view camera setup and LiDAR) to construct high definition maps.
These methods extract features from the sensor data and transform them into a Bird's Eye View (BEV) representation and derive maps in polyline representation using transformer-based architectures.
However, the construction of high quality maps requires accurate labels, but due to road construction or human error in labeling, the quality can not always guaranteed.

The goal of this thesis is to investigate the effect of map label noise on the performance of map construction models.
Therefore, different types of map label noise should be implemented in the label generation process of the dataset Argoverse 2 (https://www.argoverse.org/av2.html).
Afterwards the map construction model MapTRv2 should be trained with noisy labels and the performance evaluated on a validation dataset without map label noise.

Requirement

Requirements for students

• Knowledge in Python, PyTorch and Deep Learning
• Knowledge in Linux and Maps is a plus
• Independent working style and interest in learning new things

Faculty departments

• Engineering sciences
  Electrical engineering & information technologies
  Informatics
  Mechanical engineering
  Mechatronics & information technologies
  Mechanical Engineering

Supervision

Title, first name, last name

Jonas, Merkert

Organizational unit

Institut für Mess- und Regelungstechnik

Email address

jonas.merkert@kit.edu

Link to personal homepage/personal page

Website

Application via email

Application documents

• Curriculum vitae
• Grade transcript
• Certificate of enrollment

E-Mail Address for application
Senden Sie die oben genannten Bewerbungsunterlagen bitte per Mail an jonas.merkert@kit.edu


Back