INTRODUCING
Key to Precision and Efficiency in AI/ML Model Training
In the dynamic realm of machine learning (ML) and Artificial Intelligence (AI), GENIE emerges as a trailblazing tool by AvaWatz, crafted to redefine the approach to AI training. This comprehensive solution tackles the fundamental challenges in preparing datasets for machine learning, offering a suite of features that enhance efficiency, accuracy, reliability, and trustworthiness in ML model training.
Case Study 4
Enhancing Obstacle Detection for Tethered Drones on Roads with GENIE
Background
For tethered drones navigating roads, detecting obstacles such as traffic signal poles, electric lines, telephone poles, road signs, and bridges is crucial for safe operation. Traditional methods of training obstacle detection models are often resource-intensive and time-consuming. GENIE was employed to optimize this process.
Challenge
The challenge was to develop a machine learning model capable of identifying various classes of obstacles for a tethered drone navigating roads. The key was to create a highly accurate and diverse dataset for training the model, typically demanding extensive manual effort and time.
GENIE’s Approach
GENIE’s strategy involved a combination of representative sampling for initial labeling, automatic labeling, and advanced data augmentation techniques.
- Representative Sampling for Initial Labeling: A set of 300 samples was hand-labeled, chosen for their diversity and representativeness to cover various obstacle types.
- Automatic Labeling: GENIE automatically labeled additional images, expanding the dataset to over 2000 images with minimal manual effort.
- Data Augmentation: Techniques like copy-paste augmentation were employed to enhance the dataset’s diversity and robustness, crucial for the model’s accuracy in real-world scenarios.