The focus is on obtaining 2D and 3D labels, as well as track IDs for objects on the road. The open source, web-based 3D Bounding Box Annotation Toolbox incorporates several smart features to improve usability and efficiency. For example, semi-automatic labeling of tracks using interpolation, which is vital for downstream tasks like tracking, motion planning and motion prediction. Moreover, annotations for all camera images are automatically obtained by projecting annotations from 3D space into the image domain. In addition to the raw image and point cloud feeds, a Masterview consisting of the top view (bird’s-eye-view), side view and front view is made available to observe objects of interest from different perspectives. Comparisons of method with other publicly available annotation tools reveal that 3D annotations can be obtained faster and more efficiently by using the 3D BAT toolbox.
Calculates the optimum usage of a shipping space using Genetic Algorithms. The objective of the optimization is the maximization of loaded container of type dangerous and the minimization of the unused shipping space. The minimization of the unused shipping space has to be preferred against the maximization of the loaded container of type dangerous. A statistical evalutation can be provided.
Having more than 10 years of experience in Software Development, I am confident with Software Architecture, Object Oriented Design, Design Patterns, Test Driven Development (TDD), Continuous Integration and Delivery (CI/CD).
I am experienced Machine Learning research engineer (>4 years of experience). Developing Deep Neural Network (DNN) architectures, optimizing algorithms through hyperparameter tuning, training and evaluating machine learning models is one of my key strengths.
Experienced with data science techniques from storage of Big Data (mongoDB, MySQL), transforming data (Apache Spark), building models (Python, Pandas, SkLearn) to the visualization of data (d3.js).
I have more than 3 years of experience in developing autonomous driving software from perception (3D Object Detection) to control (MPC controller). I am expert in working with sensor data, creating ground truth datasets,
I can quickly analyze new information, find solutions and make decisions.
I am very passioned about the work I do which makes me to deliver great results in a short time.
I have experience in leading projects at university and acquired my skills in seminars and university certification programs.
Walter Zimmer, Akshay Rangesh and Mohan Trivedi
Walter Zimmer, Carsten Mueller
1. Place at one of 10 challenges, 121 teams. Road Rendering Challenge Winner Award from Autonomous Intelligent Driving (AID). Price includes visiting a conference.
Within the best 28% of master students (after 3rd master semester). Passed 7 exams (+41 ECTS) in addition.
Research stay abroad at the University of California, UCSD, San Diego, USA (16% selection rate)
Within the best 10-20% of students (same year of graduation)
Within the best 20% of students (same year of graduation)
Presenting research work (3D Labeling Tool) at the IV2019 conference
Certified in entrepreneurship.
Certified in process and quality improvement.
Training for becoming a leader (know your strengths)
Training for acquiring leadership skills
Build relationship, train intercultural communication, lead difficult conversations
DAAD English language certificate, C1 level
Build intercultural communication skills
TUM Hackathon (HackaTUM). Develop and present a prototype to several companies. Smart Garage Door (2016), roads.sexy (2019, winner)
International semester in the area of Automation Systems Engineering. Focus on Extended Concepts in Automation, Simulative Engineering, Embedded Systems. Passed 12 exams.
Certified trainer (passed two exams).
Certified Java Application Developer.
VDI is a strong network of engineers that exists for more than 160 years.
IEEE is the world largest technical professional organization for the advancement of technology.
+49 176 64372362
+49 176 64372362