Automated Focal Loss for Image based Object Detection
This addresses the time-consuming hyperparameter tuning for object detection practitioners, though it is incremental as it builds on existing focal loss methods.
The paper tackles the problem of imbalanced training data in object detection by introducing an automated focal loss that eliminates a hyperparameter, achieving up to 30% faster training convergence on COCO and up to 1.8 AOS improvement in 3D vehicle detection.
Current state-of-the-art object detection algorithms still suffer the problem of imbalanced distribution of training data over object classes and background. Recent work introduced a new loss function called focal loss to mitigate this problem, but at the cost of an additional hyperparameter. Manually tuning this hyperparameter for each training task is highly time-consuming. With automated focal loss we introduce a new loss function which substitutes this hyperparameter by a parameter that is automatically adapted during the training progress and controls the amount of focusing on hard training examples. We show on the COCO benchmark that this leads to an up to 30% faster training convergence. We further introduced a focal regression loss which on the more challenging task of 3D vehicle detection outperforms other loss functions by up to 1.8 AOS and can be used as a value range independent metric for regression.