Can I read Reliability and Calibration Analysis of Deep Object Detectors Under Extreme Class Imbalance in Safety-Critical Scenarios on EtoBox?

Reliability and Calibration Analysis of Deep Object Detectors Under Extreme Class Imbalance in Safety-Critical Scenarios by Ajeem khan is a document available to read on EtoBox.

What is Reliability and Calibration Analysis of Deep Object Detectors Under Extreme Class Imbalance in Safety-Critical Scenarios about?

This project report analyzes the reliability and calibration of deep object detectors in the context of extreme class imbalance, particularly for fire detection in safety-critical scenarios. It investigates the impact of class imbalance on detection performance metrics and confidence calibration, while evaluating various loss functions and their effects on rare-event detection. The study emphasizes the importance of lightweight models for edge deployment, addressing real-world constraints and the need for r

Author

Ajeem khan

Language

EN