Surveillance Video Synopsis via Scaling Down Objects

Abstract

Video synopsis is an effective technique to provide a compact representation of the original video by removing spatiotemporal redundancies and by preserving the essential activities. Most current approaches for video synopsis will cause collisions among objects, especially when the video is condensed much. In this paper, we present an approach for video synopsis to reduce the collisions. Our approach first shifts active objects along the time axis to compact the original video. Then, the sizes of the objects are reduced when collisions occur. Meanwhile, the geometric centroids of the objects will be kept unchanged to preserve the location information. Our contributions are threefold. First, an approach is proposed to decrease collisions in the synopsis video through reducing the sizes of the objects. Second, an optimization framework is developed to indicate the optimal time position and the appropriate reduction coefficient for each object. Finally, some metrics are proposed, and several experiments are carried out to evaluate the proposed approach. The experiments have demonstrated that the synopsis video produced by our approach has much fewer collisions while the compression ratio is high.

Video synopsis generation using spatio-temporal groups

“…Fig. 9 depicts a set of frames related to two different synopsis outputs obtained by the algorithm proposed in [4], [8], [9], [11], [12] and using our proposed method. Frames that are marked red show sudden appearance and disappearance of objects in the synopsis generated using the method proposed in [4].…”

“…It has been observed that, when E(M ) is low by grouping objects (second row), it produces better synopsis outputs. It has been observed that when the synopsis is generated using [4], [8], [9], [11], [12], due to high E(M ), the outputs are often discrete and confusing as compared to our method. We have restricted shifting of objects by grouping, therefore, our method produces longer synopsis with a lower energy difference.…”

“…Short duration tube or activity based synopsis generation described in [4], [8], [9], [12] and [11] often produces poor synopsis. It can produce synopsis with shorter lengths by showing simultaneously multiple activity segment of same object, which is confusing.…”

“…Results produced by their method produces confusing synopsis as several activities are shown simultaneously. Xuelong et al [11] have proposed methodology to minimize object collision with minimum time shifting by scaling down the object size. Pritch et al [9] have proposed method by clustering moving objects and showing similar activity together to deal with the confusing synopsis.…”

Motion Blur Detection With an Indicator Function for Surveillance Machines

“…Motion and motion blur are important information that can be used for a machine and robot, especially a surveillance machine/robot [1], to detecting the objects and products [2]- [10]. This paper focuses on a challenging and practical problem: detecting motion blur from a single image captured by a camera installed on a machine.…”