Meta-Cognition-Based Simple And Effective Approach To Object Detection
This work provides an incremental improvement for practitioners using object detection models where maintaining high speed is critical, such as in autonomous navigation.
This paper addresses the trade-off between speed and accuracy in object detection models by proposing a meta-cognitive learning strategy. This approach selectively samples training data to improve generalization, resulting in an absolute precision improvement of 2.6% to 4.4% on the MS COCO dataset with no increase in inference time.
Recently, many researchers have attempted to improve deep learning-based object detection models, both in terms of accuracy and operational speeds. However, frequently, there is a trade-off between speed and accuracy of such models, which encumbers their use in practical applications such as autonomous navigation. In this paper, we explore a meta-cognitive learning strategy for object detection to improve generalization ability while at the same time maintaining detection speed. The meta-cognitive method selectively samples the object instances in the training dataset to reduce overfitting. We use YOLO v3 Tiny as a base model for the work and evaluate the performance using the MS COCO dataset. The experimental results indicate an improvement in absolute precision of 2.6% (minimum), and 4.4% (maximum), with no overhead to inference time.