Multi-Frame Motion Detection for Active/Unstable Cameras

Network cameras, extensively used in video surveillance, often allow pan-tilt-zoom functionality and are also subject to wind load and mount vibrations, thus causing video frame misalignment. Although algorithms for motion detection, a basic block of most visual surveillance systems, are relatively mature for fixed cameras, they usually perform poorly for active and/or vibrating cameras. The issue is particularly severe for algorithms using multiple video frames jointly. In this paper, we extend our earlier work on multiple-frame motion detection to the case of active and unstable cameras. Our method accounts for spatially-affine, inter-frame transformations that can vary in time, uses a variational formulation and applies a level-set solution. We present ground-truth and real-data experimental results and show significant improvements over earlier methods.