Boosting-like Deep Learning For Pedestrian Detection

TL;DR

This paper introduces a boosting-like deep learning framework for pedestrian detection that mitigates overfitting by weighting training samples, resulting in improved performance over existing methods on benchmark datasets.

Contribution

It presents a novel boosting-inspired technique integrated into deep learning to address overfitting in pedestrian detection tasks.

Findings

Achieves 15.85% reduction in miss rate compared to ACF

Achieves 3.81% reduction in miss rate compared to JointDeep

Demonstrates stable and superior performance on Caltech dataset

Abstract

This paper proposes boosting-like deep learning (BDL) framework for pedestrian detection. Due to overtraining on the limited training samples, overfitting is a major problem of deep learning. We incorporate a boosting-like technique into deep learning to weigh the training samples, and thus prevent overtraining in the iterative process. We theoretically give the details of derivation of our algorithm, and report the experimental results on open data sets showing that BDL achieves a better stable performance than the state-of-the-arts. Our approach achieves 15.85% and 3.81% reduction in the average miss rate compared with ACF and JointDeep on the largest Caltech benchmark dataset, respectively.