ViS-HuD: Using Visual Saliency to Improve Human Detection with Convolutional Neural Networks

TL;DR

The paper introduces ViS-HuD, a novel deep learning approach that enhances human detection in images by integrating visual saliency maps with CNNs, achieving state-of-the-art accuracy on benchmark datasets.

Contribution

It proposes a new method combining visual saliency maps with CNNs for improved human detection in images, leveraging pre-trained models ML-Net and DetectNet.

Findings

Achieves 91.4% detection accuracy on Penn Fudan dataset.

Attains 53% average miss-rate on TUD-Brussels benchmark.

Outperforms existing methods in human detection performance.

Abstract

The paper presents a technique to improve human detection in still images using deep learning. Our novel method, ViS-HuD, computes visual saliency map from the image. Then the input image is multiplied by the map and product is fed to the Convolutional Neural Network (CNN) which detects humans in the image. A visual saliency map is generated using ML-Net and human detection is carried out using DetectNet. ML-Net is pre-trained on SALICON while, DetectNet is pre-trained on ImageNet database for visual saliency detection and image classification respectively. The CNNs of ViS-HuD were trained on two challenging databases - Penn Fudan and TUD-Brussels Benchmark. Experimental results demonstrate that the proposed method achieves state-of-the-art performance on Penn Fudan Dataset with 91.4% human detection accuracy and it achieves average miss-rate of 53% on the TUDBrussels benchmark.