How Far are We from Solving Pedestrian Detection?

TL;DR

This paper analyzes the gap between current pedestrian detection methods and the ideal detector, identifying key error types and proposing improvements through data sanitization and convnet analysis.

Contribution

It introduces a human baseline for pedestrian detection, analyzes recurrent errors, and explores the effects of annotation noise and convnet factors on detection performance.

Findings

Identified localization and background-foreground errors in pedestrian detection

Improved detection performance with sanitized training data

Achieved top results on the Caltech dataset

Abstract

Encouraged by the recent progress in pedestrian detection, we investigate the gap between current state-of-the-art methods and the "perfect single frame detector". We enable our analysis by creating a human baseline for pedestrian detection (over the Caltech dataset), and by manually clustering the recurrent errors of a top detector. Our results characterize both localization and background-versus-foreground errors. To address localization errors we study the impact of training annotation noise on the detector performance, and show that we can improve even with a small portion of sanitized training data. To address background/foreground discrimination, we study convnets for pedestrian detection, and discuss which factors affect their performance. Other than our in-depth analysis, we report top performance on the Caltech dataset, and provide a new sanitized set of training and test annotations.