Just Add $ 100 More: Augmenting Pseudo-LiDAR Point Cloud for Resolving Class-imbalance Problem
–Neural Information Processing Systems
Typical LiDAR-based 3D object detection models are trained with real-world data collection, which is often imbalanced over classes. To deal with it, augmentation techniques are commonly used, such as copying ground truth LiDAR points and pasting them into scenes.
Neural Information Processing Systems
May-30-2025, 06:36:17 GMT
- Country:
- Asia (1.00)
- Europe (1.00)
- North America > United States
- California (0.46)
- Genre:
- Research Report
- Experimental Study (0.93)
- New Finding (0.67)
- Research Report
- Industry:
- Information Technology (1.00)
- Media > Film (0.46)
- Technology:
- Information Technology > Artificial Intelligence
- Machine Learning > Neural Networks (0.68)
- Representation & Reasoning (1.00)
- Robots > Autonomous Vehicles (0.46)
- Vision (1.00)
- Information Technology > Artificial Intelligence