Search

Salient object detection is a fundamental problem in computer vision. Existing methods using only low-level features failed to uniformly highlight the salient object regions. In order to combine high-level saliency priors and low-level appearance cues, we propose a novel Background Prior based Salient detection method (BPS) for high-quality salient object detection. Different from other background prior based methods, a background estimation is added before performing saliency detection. We utilize the distribution of bounding boxes generated by a generic object proposal method to obtain background information. Three background priors are mainly considered to model the saliency, namely background connectivity prior , background contrast prior and spatial distribution prior , allowing the proposed method to highlight the salient object as a whole and suppress background clutters. Experiments conducted on two benchmark datasets validate that our method outperforms 11 state-of-the-art methods, while being more efficient than most leading methods. [ABSTRACT FROM AUTHOR]

Highlights: [•] A new common visual patterns (CVPs) discovery framework is proposed. [•] It contains three successive steps: initialization, expansion and combination. [•] It can effectively deal with image photometric and geometric transformations. [•] The CVPs can be used for object recognition and near-duplicate image retrieval. [ABSTRACT FROM AUTHOR]

Abstract: The issue of personal privacy has garnered significant attention with the extensive application of video surveillance systems. Privacy region scrambling is an effective method to protect privacy in video. To ensure that nonprivacy regions are not affected by scrambling, particular methods must be taken to prevent drift error in privacy protected video scrambling. However, existing methods have significantly reduced the coding efficiency. In this paper, we focus on improving coding efficiency while preventing drift error in privacy protected H.264/AVC video scrambling, which is the state-of-the-art coding standard. A restricted video coding scheme is proposed, which involves three parts of Mode Restricted Intra Prediction (MRIP), Search Window Restricted Motion Estimation (SWRME) and Boundary Strength Restricted Deblocking Filtering (BSRDF). Experimental results show that the proposed restricted video coding scheme prevents drift error with higher coding efficiency than others. [Copyright &y& Elsevier]

A new integer 4×4 transform is adopted in the new video coding standard H.264/AVC. It consumes a big account of computation of a real-time encoder. To reduce the computation load of the new transform, pruned DCT algorithm is applied. To avoid serious encoding performance loss, a new condition is derived based on statistical sum of absolute value test(SSAVT) to identify pruned block adaptively. Considering the variance of DCT coefficients’ contribution, a novel efficient model based on statistical analysis is proposed. To achieve further control on DCT complexity and a good tradeoff between performance and computation load, a scheme to control the complexity of DCT is presented. Simulation results show that the novel statistical model can save more computation, so it is chose to control DCT complexity. Simulation results show that the complexity controllable scheme can control the complexity of DCT within the specified target with negligible encoding performance loss. [Copyright &y& Elsevier]