Preserving interactions among moving objects in surveillance video synopsis
- PDF / 10,399,927 Bytes
- 30 Pages / 439.642 x 666.49 pts Page_size
- 87 Downloads / 223 Views
Preserving interactions among moving objects in surveillance video synopsis Namitha K1
· Athi Narayanan1
Received: 1 October 2019 / Revised: 10 July 2020 / Accepted: 29 July 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Video synopsis is an effective solution for fast browsing and retrieval of long surveillance videos. It aims to shorten long video sequences into its equivalent compact video representation by rearranging the video events in the temporal domain and/or spatial domain. Conventional video synopsis methods focus on reducing the collisions between tubes and maintaining their chronological order, which may alter the original interactions between tubes due to improper tube rearrangement. In this paper, we present an approach to preserve the relationships among tubes (tracks of moving objects) of the original video in the synopsis video. First, a recursive tube-grouping algorithm is proposed to determine the behavior interactions among tubes in a video and group the related tubes together to form tube sets. Second, to preserve the discovered relationships, a spatio-temporal cube voting algorithm is proposed. This cube voting method optimally rearranges the tube sets in the synopsis video, minimizing false collisions between tubes. Third, a method to estimate the duration of the synopsis video is proposed based on an entropy measure of tube collisions. The extensive experimental results demonstrate that the proposed video synopsis framework condenses videos by preserving the original tube interactions and reducing false tube collisions. Keywords Video synopsis · Surveillance · Interaction · Tube grouping · Tube rearrangement
1 Introduction With the growing demand for security solutions in public and private sectors, surveillance cameras play a huge role in our day-to-day lives. The enormous amount of videos captured by these cameras are increasing explosively, creating challenges in its effective retrieval and review. In most cases, surveillance footage contains highly redundant data with only lim Namitha K
[email protected] Athi Narayanan [email protected] 1
Department of Computer Science and Engineering, Amrita Vishwa Vidyapeetham, Amritapuri, India
Multimedia Tools and Applications
ited useful information. Therefore, it is time-consuming to browse through such massive video data for inspecting an event of interest. Several techniques such as video fast forward [17], video skimming [30], video montage [21], video condensation [25] and video summarization [8, 12, 26] have been proposed to address the aforementioned challenge. However, these methods lose the dynamic aspect of video or have apparent stitching seams in the summary. To address these drawbacks, an object-based technique, video synopsis [38, 39, 41] has been proposed in the literature. Video synopsis provides a compact video representation of long surveillance videos where multiple objects are displayed simultaneously, irrespective of their time of occurrence. The fundamental processing bl
Data Loading...
