Reduced Memory Region Based Deep Convolutional Neural Network Detection

TL;DR

This paper presents a memory-efficient CNN-based pedestrian detector optimized for embedded systems, achieving high accuracy with low computational and memory requirements.

Contribution

It demonstrates that a region-based deep neural network can be finely tuned for pedestrian detection and compressed to operate effectively on low-memory devices.

Findings

Outperforms traditional feature-based methods in accuracy

Achieves near state-of-the-art accuracy with low computational complexity

Maintains detection performance after model compression

Abstract

Accurate pedestrian detection has a primary role in automotive safety: for example, by issuing warnings to the driver or acting actively on car's brakes, it helps decreasing the probability of injuries and human fatalities. In order to achieve very high accuracy, recent pedestrian detectors have been based on Convolutional Neural Networks (CNN). Unfortunately, such approaches require vast amounts of computational power and memory, preventing efficient implementations on embedded systems. This work proposes a CNN-based detector, adapting a general-purpose convolutional network to the task at hand. By thoroughly analyzing and optimizing each step of the detection pipeline, we develop an architecture that outperforms methods based on traditional image features and achieves an accuracy close to the state-of-the-art while having low computational complexity. Furthermore, the model is compressed in order to fit the tight constrains of low power devices with a limited amount of embedded memory available. This paper makes two main contributions: (1) it proves that a region based deep neural network can be finely tuned to achieve adequate accuracy for pedestrian detection (2) it achieves a very low memory usage without reducing detection accuracy on the Caltech Pedestrian dataset.