Fast Appearance Modeling for Automatic Primary Video Object Segmentation

—Automatic segmentation of the primary object in a video clip is a challenging problem as there is no prior knowledge of the foreground object. Most existing techniques thus adapt an iterative approach for foreground and background appearance modeling, i.e., fix the appearance model while optimizing the segmentation and fix the segmentation while optimizing the appearance model. However, these approaches may rely on good initiation and can be easily trapped at local optimal. Also, they are usually time consuming for analyzing videos. To address these limitations, we propose a novel and efficient appearance modeling technique for automatic primary video object segmentation in the Markov Random Field (MRF) framework. It embeds the appearance constraint as auxiliary nodes and edges in the MRF structure, and can optimize both the segmentation and appearance model parameters simultaneously in one graph cut. Extensive experimental evaluations validate the superiority of the proposed met...

• Type: Journal
• Year: 2016
• Where: TIP
• Authors: Jiong Yang, Brian L. Price, Xiaohui Shen, Zhe L. Lin, Junsong Yuan
• Keywords: Computer Science, TIP 2016