Object Detection with Pixel Intensity Comparisons Organized in Decision Trees

TL;DR

This paper introduces a fast object detection method using decision trees with pixel intensity comparisons, organized in a cascade, demonstrating practical face detection with noise robustness and rotation invariance.

Contribution

It presents a novel, efficient decision tree-based approach for object detection that is fast, noise-resistant, and capable of rotation invariance, with open-source implementation.

Findings

Effective face detection results.

High processing speed due to pixel intensity comparisons.

Robustness to noise and rotation.

Abstract

We describe a method for visual object detection based on an ensemble of optimized decision trees organized in a cascade of rejectors. The trees use pixel intensity comparisons in their internal nodes and this makes them able to process image regions very fast. Experimental analysis is provided through a face detection problem. The obtained results are encouraging and demonstrate that the method has practical value. Additionally, we analyse its sensitivity to noise and show how to perform fast rotation invariant object detection. Complete source code is provided at https://github.com/nenadmarkus/pico.