Search

We propose a novel robust hashing algorithm based on signal-to-noise ratio (SNR) maximization to learn compact binary codes, where the SNR metric is used to select a set of projection directions, and one hash bit is extracted from each projection direction. We first motivate this approach under a Gaussian model for the underlying signals, in which case maximizing SNR is equivalent to minimizing the robust hashing error probability. A globally optimal solution can be obtained by solving a generalized eigenvalue problem. We also develop a multibit per projection algorithm to learn longer hash codes when the number of high-SNR projections is limited. The proposed algorithms are tested on both synthetic and real data sets, showing significant performance gains over existing hashing algorithms. [ABSTRACT FROM PUBLISHER]

Sphingosine-1-phosphate (SIP) lyase is considered as a drug target in autoimmune diseases based on the protective effect of reducing activity of the enzyme in animal models of inflammation. Since SIP lyase deficiency in mice causes a severe, lethal phenotype, it was of interest to investigate any pathological alterations associated with only partially reduced activity of SIP lyase as may be encountered upon pharmacological inhibition. Both genetic reduction of SIP lyase activity in mice and inhibition of SIP lyase with a low-molecular-weight compound in rats consistently resulted in podocyte-based kidney toxicity, which is the most severe finding. In addition, skin irritation and platelet activation were observed in both instances. The similarity of the findings in both the genetic model and the pharmacological study supports the value of analyzing inducible partially target-deficient mice for safety assessment. If the findings described in rodents translate to humans, target-related toxicity, particularly podocyte dysfunction, may limit chronic systemic treatment of autoimmune diseases with SIP lyase inhibitors. Furthermore, partial deficiency or inhibition of SIP lyase appears to provide an in vivo rodent model to enable studies on the mechanism of podocyte dysfunction. [ABSTRACT FROM AUTHOR]

Ineffective spatial-temporal motion feature pooling has been a fundamental bottleneck for human action recognition/detection for decades. Previous pooling schemes such as global, spatial-temporal pyramid, or human and object centric pooling fail to capture discriminative motion patterns because informative movements only occur in specific regions of the human body, that depend on the type of action being performed. Global (holistic) motion feature pooling methods therefore often result in an action representation with limited discriminative capability. To address this fundamental limitation, we propose an adaptive motion feature pooling scheme that utilizes human poses as side information. Such poses can be detected for instance in assisted living and indoor smart surveillance scenarios. Taking both video sub-volumes for pooling and human pose types as hidden variables, we formulate the motion feature pooling problem as a latent structural learning problem where the relationship between the discriminative pooling video sub-volumes and the pose types is learned. The resulting pose adaptive motion feature pooling scheme is extensively tested on assisted living and smart surveillance datasets and on general action recognition benchmarks. Improved action recognition and detection performances are demonstrated. [ABSTRACT FROM AUTHOR]