Yuanfang Guo

Image Deblurring via Extreme Channels Prior

Camera motion introduces motion blur, affecting many computer vision tasks. Dark Channel Prior (DCP) helps the blind deblurring on scenes including natural, face, text, and low-illumination images. However, it has limitations and is less likely to support the kernel estimation while bright pixels dominate the input image. We observe that the bright pixels in the clear images are not likely to be bright after the blur process. Based on this observation, we first illustrate this phenomenon mathematically and define it as the Bright Channel Prior (BCP). Then, we propose a technique for deblurring such images which elevates the performance of existing motion deblurring algorithms. The proposed method takes advantage of both Bright and Dark Channel Prior. This joint prior is named as extreme channels prior and is crucial for achieving efficient restorations by leveraging both the bright and dark information. Extensive experimental results demonstrate that the proposed method is more robust and performs favorably against the state-of-the-art image deblurring methods on both synthesized and natural images.

Binarized Mode Seeking for Scalable Visual Pattern Discovery

This paper studies visual pattern discovery in large-scale image collections via binarized mode seeking, where images can only be represented as binary codes for efficient storage and computation. We address this problem from the perspective of binary space mode seeking. First, a binary mean shift (bMS) is proposed to discover frequent patterns via mode seeking directly in binary space. The binomial-based kernel and binary constraint are introduced for binarized analysis. Second, we further extend bMS to a more general form, namely contrastive binary mean shift (cbMS), which maximizes the contrastive density in binary space, for finding informative patterns that are both frequent and discriminative for the dataset. With the binarized algorithm and optimization, our methods demonstrate significant computation (50×) and storage (32×) improvement compared to standard techniques operating in Euclidean space, while the performance does not largely degenerate. Furthermore, cbMS discovers more informative patterns by suppressing low discriminative modes. We evaluate our methods on both annotated ILSVRC (1M images) and un-annotated blind Flickr (10M images) datasets with million scale images, which demonstrates both the scalability and effectiveness of our algorithms for discovering frequent and informative patterns in large scale collection.

An Adaptive Reversible Data Hiding Scheme for JPEG Images

Currently JPEG is the most popular image file format and the majority of images are stored in JPEG format due to storage constraint. Recently, reversible data hiding (RDH) for JPEG images draws researchers attention and has been developed rapidly. Due to the compression, performing RDH on a typical JPEG image is much more difficult than that on an uncompressed image. In this paper, we propose an adaptive reversible data hiding method for JPEG images, which is based on histogram shifting. We propose to select the optimal expandable bins-pair at image level by adopting a k-th nearest neighbors (KNN) algorithm. By developing a new block selection strategy, we can adaptively select the to-be-embedded blocks. Then, the message bits are embedded into the selected blocks at a specific bins-pair via the histogram shifting algorithm. Experimental results demonstrate that our proposed method can achieve a higher image quality and a less increased file size compared to the current state-of-the-art RDH method for JPEG images.

3-in-1 Correlated Embedding via Adaptive Exploration of the Structure and Semantic Subspaces

Combinational network embedding, which learns the node representation by exploring both topological and non-topological information, becomes popular due to the fact that the two types of information are complementing each other. Most of the existing methods either consider the topological and non-topological information being aligned or possess predetermined preferences during the embedding process. Unfortunately, previous methods fail to either explicitly describe the correlations between topological and non-topological information or adaptively weight their impacts. To address the existing issues, three new assumptions are proposed to better describe the embedding space and its properties. With the proposed assumptions, nodes, communities and topics are mapped into one embedding space. A novel generative model is proposed to formulate the generation process of the network and content from the embeddings, with respect to the Bayesian framework. The proposed model automatically leans to the information which is more discriminative. The embedding result can be obtained by maximizing the posterior distribution by adopting the variational inference and reparameterization trick. Experimental results indicate that the proposed method gives superior performances compared to the state-of-the-art methods when a variety of real-world networks is analyzed.

Contextual approach for identifying malicious Inter-Component privacy leaks in Android apps

Inter-Component Communication (ICC) enables developers to create rich and innovative applications in Android platform. However, some privacy problems occur because of the interactions among multiple components. Since the flow of sensitive data across components may be legal or malicious, it is necessary to perform a precise ICC analysis to identify the malicious flow of sensitive data. In this paper, we propose a static taint analysis method, named IccChecker, to identify the malicious ICC-based privacy leaks in Android applications. IccChecker first tracks the potential flow of sensitive data across components and extracts the contextual factors which trigger the sensitive behavior. By leveraging the context information, our approach differentiates the malicious privacy leaks from the legal privacy information exchanges according to the proposed contextual policy. Moreover, we present a comprehensive assessment with benchmarks and real-world applications. Our evaluation results with benchmarks demonstrate that IccChecker improves the precision of ICC-based privacy leak detection. In the evaluation with real-world applications, our approach identifies 4 apps with ICC-based privacy leaks among 168 Google Play apps (2.3%) while 31 apps are identified from 49 malwares (63.3%).