Scale Invariant Mask R-CNN for Pedestrian Detection

Abstract

Pedestrian detection is a challenging and active research area in computer vision. Recognizing pedestrians helps in various utility applications such as event detection in overcrowded areas, gender, and gait classification, etc. In this domain, the most recent research is based on instance segmentation using Mask R-CNN. Most of the pedestrian detection method uses a feature of different body portions for identifying a person. This feature-based approach is not efficient enough to differentiate pedestrians in real-time, where the background changing. In this paper, a combined approach of scale-invariant feature map generation for detecting a small pedestrian and Mask R-CNN has been proposed for multiple pedestrian detection to overcome this drawback. The new database was created by recording the behavior of the student at the prominent places of the engineering institute. This database is comparatively new for pedestrian detection in the academic environment. The proposed Scale-invariant Mask R-CNN has been tested on the newly created database and has been compared with the Caltech [1], INRIA [2], MS COCO [3], ETH [4], and KITTI [5] database. The experimental result shows significant performance improvement in pedestrian detection as compared to the existing approaches of pedestrian detection and instance segmentation. Finally, we conclude and investigate the directions for future research.