3D Trajectory Reconstruction of a Moving Object from a Stereo Video using Particle Filter

Abstract

Moving object detection and tracking in videos is an actively researched area for the last two decades due to
its practical applications in many areas, such as trajectory analysis of moving objects, making game playing robots,
human computer interaction, etc. This paper presents an experimental study to reconstruct the 3D trajectory of a
coloured moving object by combing particle filter and stereo vision. Although, the initial results obtained are
encouraging, but the generated trajectory has waviness. Further research is required to reduce the waviness and
generate more accurate trajectory.