Sampling Strategies for Efficient Training of Deep Learning Object Detection Algorithms

TL;DR

This paper investigates two sampling strategies, uniform and frame difference sampling, to improve training efficiency of deep learning object detection models by reducing the need for manual labels while maintaining performance.

Contribution

It introduces and evaluates two novel sampling strategies based on Lipschitz continuity assumptions to enhance training efficiency in deep learning object detection.

Findings

Both strategies achieve comparable performance with fewer labeled samples.

Frame difference sampling effectively exploits temporal redundancy in videos.

Uniform sampling ensures diverse coverage of the state space.

Abstract

Two sampling strategies are investigated to enhance efficiency in training a deep learning object detection model. These sampling strategies are employed under the assumption of Lipschitz continuity of deep learning models. The first strategy is uniform sampling which seeks to obtain samples evenly yet randomly through the state space of the object dynamics. The second strategy of frame difference sampling is developed to explore the temporal redundancy among successive frames in a video. Experiment result indicates that these proposed sampling strategies provide a dataset that yields good training performance while requiring relatively few manually labelled samples.