Abstract
Today industries strive toward using data-driven
machine learning wherever applicable. Consequently, they re-
quire manually or automatically labeled training data sets. Currently, synthetically generating labeled training data sets belongs
to the open challenges in machine learning across multiple application fields. In this paper, we propose employing a procedural
pipeline combining BlenderProc with domain randomization to
create prelabeled training data sets synthetically. Randomizing
the domain using uncorrelated random background images, we
ensure that the neural network applied for object detection purely
learns the object features and is background-independent. Our
proposed pipeline yields a solution to create sizeable prelabeled
training data sets. We assess the pipeline performance for the
application of cone object detection for the formula student
driverless competition using no real training and a small real-world training data set for fine-tuning: We show that using the
synthetically generated training data fine-tuned with a limited
real training data set performs best for object detection. This
transfer learning-based, fine-tuned solution also outperforms the
benchmark training data set in detecting knocked-over cones
that are neither present in the real nor the synthetic training
data set. Consequently, by combining BlenderProc and domain
randomization, we provide a solution for formula student teams
to generate extensive training data for cone detection and other
detection problems relevant to driverless.
machine learning wherever applicable. Consequently, they re-
quire manually or automatically labeled training data sets. Currently, synthetically generating labeled training data sets belongs
to the open challenges in machine learning across multiple application fields. In this paper, we propose employing a procedural
pipeline combining BlenderProc with domain randomization to
create prelabeled training data sets synthetically. Randomizing
the domain using uncorrelated random background images, we
ensure that the neural network applied for object detection purely
learns the object features and is background-independent. Our
proposed pipeline yields a solution to create sizeable prelabeled
training data sets. We assess the pipeline performance for the
application of cone object detection for the formula student
driverless competition using no real training and a small real-world training data set for fine-tuning: We show that using the
synthetically generated training data fine-tuned with a limited
real training data set performs best for object detection. This
transfer learning-based, fine-tuned solution also outperforms the
benchmark training data set in detecting knocked-over cones
that are neither present in the real nor the synthetic training
data set. Consequently, by combining BlenderProc and domain
randomization, we provide a solution for formula student teams
to generate extensive training data for cone detection and other
detection problems relevant to driverless.
| Original language | English |
|---|---|
| Title of host publication | 2022 International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME) |
| Number of pages | 6 |
| Publisher | IEEE |
| Publication date | 2022 |
| ISBN (Electronic) | 978-1-6654-7095-7 |
| DOIs | |
| Publication status | Published - 2022 |
| Event | 2022 International Conference on Electrical, Computer, Communications and Mechatronics Engineering 2022 - Duration: 16. Nov 2022 → 18. Nov 2022 http://www.iceccme.com/ |
Conference
| Conference | 2022 International Conference on Electrical, Computer, Communications and Mechatronics Engineering 2022 |
|---|---|
| Period | 16/11/2022 → 18/11/2022 |
| Internet address |
Keywords
- Synthetic Training Data
- Domain Randomization
- BlenderProc
- Driverless
- Transfer Learning
- Formula Student
- transfer learning
- driverless
- formula student
- synthetic training data
- domain randomization