An Occlusion Reasoning Scheme for Monocular Pedestrian Tracking in Dynamic Scenes

TL;DR

This paper introduces an occlusion reasoning scheme for monocular pedestrian tracking in dynamic scenes, using binary integer programming and SURF matching to improve association accuracy amidst occlusions and camera motion.

Contribution

It proposes a novel frame-by-frame association method combining affinity matrices, binary integer programming, and SURF verification for robust pedestrian tracking.

Findings

Effective in handling occlusions and camera motion

Improves tracking accuracy on standard datasets

Outperforms some existing methods in challenging scenarios

Abstract

This paper looks into the problem of pedestrian tracking using a monocular, potentially moving, uncalibrated camera. The pedestrians are located in each frame using a standard human detector, which are then tracked in subsequent frames. This is a challenging problem as one has to deal with complex situations like changing background, partial or full occlusion and camera motion. In order to carry out successful tracking, it is necessary to resolve associations between the detected windows in the current frame with those obtained from the previous frame. Compared to methods that use temporal windows incorporating past as well as future information, we attempt to make decision on a frame-by-frame basis. An occlusion reasoning scheme is proposed to resolve the association problem between a pair of consecutive frames by using an affinity matrix that defines the closeness between a pair of windows and then, uses a binary integer programming to obtain unique association between them. A second stage of verification based on SURF matching is used to deal with those cases where the above optimization scheme might yield wrong associations. The efficacy of the approach is demonstrated through experiments on several standard pedestrian datasets.