Superpixel-based video object segmentation using perceptual organization and location prior

Gamifying Video Object Segmentation

“…Our work shares the same end goal of automated video object segmentation [1], [2], [19], [20], [21], [22], [23], [24], but our approach is more inline with research that places humans in the loop (including games with purpose) [11], [12], [13], [14], [15], [16], [17], [25], [26], [27], [28], [29], [30], [31], [32] and interactive video segmentation [7], [33], [34], [35], [36], [37], [38], [39].…”

“…Indeed, every day several petabytes of videos are routinely generated for disparate applications ranging from videosurveillance to news broadcasting to entertainment. This is also highlighted in the recent "Forecast and Methodology 2014-2019" report 1 by CISCO that has estimated that consumer internet video traffic will be 80 percent of all consumer Internet traffic in 2019. Nevertheless, this video data deluge needs automated methods able to extract meaningful information for data indexing, preservation and understanding, since it is unrealistic and unfeasible (as stated in the same CISCO report, it would take an individual over 5 million years to watch the amount of video that will cross global IP networks each month in 2019) to assume people can do Manuscript submitted on January 05, 2016 1. http://www.cisco.com/c/en/us/solutions/collateral/serviceprovider/ip-ngn-ip-next-generation-network/white paper c11-481360.html this work manually.…”

“…Unsupervised video segmentation has gained a lot of attention in the last decades [2], [19], [20], [21], [22] and recently it has been thought mainly in terms of spatiotemporal superpixel modeling [1], [23], [24]. The key idea behind these methods is the one of grouping pixels which are appearance-and motion-wise-consistent.…”

“…There is a large (past and present) bulk of literature of methods for video object segmentation. Background modeling/subtraction [1], [2], motion analysis [3], [4], object ranking [5], [6] and clustering point tracks [7], [8] and, recently, combination of CNN-based moving object detectors [9] are among the most common methods. However, the accuracy and performance of these techniques are still not satisfactory, especially in cases of articulated motion, cluttered scenes and object occlusions.…”

Threshold Based Filtering Technique for Efficient Moving Object Detection and Tracking in Video Surveillance

“…A tracking framework introduced in [13] for tracking the multiple objects based on Continuous Energy Minimization. In [14], a new approach was presented for segmenting objects in videos taken in complex scenes with multiple and different targets.…”