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

Vandit Gajjar, Yash Khandhediya, Ayesha Gurnani, Viraj Mavani, Mehul S. Raval; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2018, pp. 1908-1916

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 TUD-Brussels benchmark.

Related Material


[pdf] [arXiv]
[bibtex]
@InProceedings{Gajjar_2018_CVPR_Workshops,
author = {Gajjar, Vandit and Khandhediya, Yash and Gurnani, Ayesha and Mavani, Viraj and Raval, Mehul S.},
title = {ViS-HuD: Using Visual Saliency to Improve Human Detection With Convolutional Neural Networks},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2018}
}