Learn More
Low-level saliency cues or priors do not produce good enough saliency detection results especially when the salient object presents in a low-contrast background with confusing visual appearance. This issue raises a serious problem for conventional approaches. In this paper, we tackle this problem by proposing a multi-context deep learning framework for(More)
Learning generic and robust feature representations with data from multiple domains for the same problem is of great value, especially for the problems that have multiple datasets but none of them are large enough to provide abundant data variations. In this work, we present a pipeline for learning deep feature representations from multiple domains with(More)
Nonvoltage-gated cation currents, which are activated following stimulation of phospholipase C (PLC), appear to be major modes for Ca2+ and Na+ entry in mammalian cells. The TRPC channels may mediate some of these conductances since their expression in vitro leads to PLC-dependent cation influx. We found that the TRPC3 protein was highly enriched in neurons(More)
Cross-scene crowd counting is a challenging task where no laborious data annotation is required for counting people in new target surveillance crowd scenes unseen in the training set. The performance of most existing crowd counting methods drops significantly when they are applied to an unseen scene. To address this problem, we propose a deep convolutional(More)
The Drosophila retinal-specific protein, TRP (transient receptor potential), is the founding member of a family of store-operated channels (SOCs) conserved from C. elegans to humans. In vitro studies indicate that TRP is a SOC, but that the related retinal protein, TRPL, is constitutively active. In the current work, we report that coexpression of TRP and(More)
The light response in Drosophila photoreceptor cells is mediated by a series of proteins that assemble into a macromolecular complex referred to as the signalplex. The central player in the signalplex is inactivation no afterpotential D (INAD), a protein consisting of a tandem array of five PDZ domains. At least seven proteins bind INAD, including the(More)
Automatically assigning relevant text keywords to images is an important problem. Many algorithms have been proposed in the past decade and achieved good performance. Efforts have focused upon model representations of keywords, whereas properties of features have not been well investigated. In most cases, a group of features is preselected, yet important(More)
Automatically assigning relevant text keywords to images is an important problem. Many algorithms have been proposed in the past decade and achieved good performance. Efforts have focused upon model representations of keywords, but properties of features have not been well investigated. In most cases, a group of features is preselected, yet important(More)
The Drosophila visual system has provided a model to study phototransduction and retinal degeneration. To identify new candidate proteins that contribute to these processes, we conducted a genome-wide screen for genes expressed predominately in the eye, using DNA microarrays. This screen appeared to be comprehensive as it led to the identification of all 22(More)
In this paper, we introduce a new matching method based on a novel locally affine-invariant geometric constraint and linear programming techniques. To model and solve the matching problem in a linear programming formulation, all geometric constraints should be able to be exactly or approximately reformulated into a linear form. This is a major difficulty(More)