Video synchronization using bit rate profiles

Abstract

We present a novel approach for the temporal synchronization of multiple videos which is based on cross-correlating bit rate profiles. The proposed scheme determines the temporal offset without major restrictions on viewing angles, camera properties and camera motion. We propose two extensions of the basic algorithm which reduce the influence of camera motion and distracting background objects. Additionally, we describe how to optimally combine different bit rate components in order to further improve the reliability of the synchronization scheme. The proposed approach, when combined with the three extensions, leads to a reliable, robust, and frame accurate temporal alignment of videos at remarkably low complexity.

Fully Automatic and Frame-Accurate Video Synchronization Using Bitrate Sequences

“…In [2], we have presented a fundamentally different approach to video synchronization. Instead of imitating the human eye in detecting synchronous events in two videos, our approach depends on a more abstract concept: the information content of individual video frames.…”

“…Starting from the initial discovery in [2] that bitrate profiles are a powerful source of information for video synchronization, we proposed a robust video synchronization algorithm based on Consensus-based Crosscorrelation (ConCor) in [3]. In this article, we extend our preliminary work from [2] and [3] in several directions. We specifically describe how to extend ConCor with a normalization scheme so as to improve the alignment as a whole.…”

“…Moreover, we analyze the re-encoding process in general and derive practical guidelines for synchronizationrelated bitrate extraction based on H.264/AVC. Compared to [2] and [3], the extensions described in this paper lead to a substantial performance improvement. For the tested datasets, in comparison with [3], the percentage of videos perfectly synchronized rises from 50% to 67%.…”

Synchronization of presentation slides and lecture videos using bit rate sequences

“…A low complexity approach for the synchronization of multiple video sequences, which is based on the correlation of bit rate profiles, has been proposed in [10]. However, slide synchronization using bit rate information has not been studied in previous work.…”

“…Objects entering the field-of-view of the camera can cause a more significant increase as new texture is introduced. However, by parsing the bit rate block by block, as proposed in [10], the contribution of boundary macro blocks, which are most affected by new emerging objects, can be ignored. Altogether, with the proposed smart bit rate extraction, the bit rate signal is robust against foreground motion, entering objects, and small camera motion (e.g.…”