Localized temporal profile of surveillance video

Abstract

Surveillance videos are recorded pervasively and their retrieval currently still relies on human operators. As an intermediate representation, this work develops a new temporal profile of video to convey accurate temporal information in the video while keeping certain spatial characteristics of targets of interest for recognition. The profile is obtained at critical positions where major target flow appears. We set a sampling line crossing the motion direction to profile passing targets in the temporal domain. In order to add spatial information to the temporal profile to certain extent, we integrate multiple profiles from a set of lines with blending method to reflect the target motion direction and position in the temporal profile. Different from mosaicing/montage methods for video synopsis in spatial domain, our temporal profile has no limit on the time length, and the created profile significantly reduces the data size for brief indexing and fast search of video.

Chronological Video Synopsis via Events Rearrangement Optimization

“…Feng et al [8] and Huang et al [9,10] proposed online temporal location strategy to keep the temporal order and reduce the computational complexity, but the collision problem remain unsolved. Lu et al [11] , Saeid Bagheri et al [12] and Zhang et al [13] made use of objects trajectory information to reduce the data size of events. Nie et al [14] presented a multilevel path relocation method, composited with global spatiotemporal optimization to deal with the objects collision.…”

Temporal mapping of surveillance video for indexing and summarization

“…Therefore, one can easily understand that, for a target moves rightward to across the sampling lines, the transparencies of shapes 20 increase in the profile. Inversely, a leftward motion generates shapes more and more opaque in the temporal profile [36]. Resulting from blending multiple slices together, this temporal profile not only allows us to locate time instances of passing targets, but also helps us understand the motion direction of targets.…”