Showing papers by "Sheng Tang published in 2014"
TL;DR: This paper proposes a novel framework of fusing audio vocabulary with visual features for pornographic video detection, and shows that this approach outperforms the traditional one which is only based on visual features, and achieves satisfactory performance.
16 citations
TL;DR: PA4 variant rs6467136 was associated with the therapeutic effect of rosiglitazone in Chinese T2DM patients, and single-nucleotide polymorphisms showed no effect on repaglinide efficacy.
Abstract: The aim of this study was to investigate the association of PAX4 variants with therapeutic effect of oral antidiabetic drugs in Chinese type 2 diabtes mellitus (T2DM) patients. A total of 209 newly diagnosed T2DM patients were randomly assigned to treatment with repaglinide or rosiglitazone for 48 weeks, and the therapeutic effects were compared. In the rosiglitazone cohort, rs6467136 GA+AA carriers showed greater decrease in 2-h glucose levels (P=0.0063) and higher cumulative attainment rates of target 2-h glucose levels (Plog rank=0.0093) than GG homozygotes. In the subgroup with defective β-cell function, rs6467136 GA+AA carriers exhibited greater decrements of 2-h glucose level and improvement of homeostasis model assessment of insulin resistance (P=0.0143). Moreover, GA+AA carriers were more likely to attain the target fasting and 2-h glucose level (Plog rank=0.0091 and 0.007, respectively). However, these single-nucleotide polymorphisms showed no effect on repaglinide efficacy. In conclusion, PAX4 variant rs6467136 was associated with the therapeutic effect of rosiglitazone in Chinese T2DM patients.
14 citations
01 Oct 2014
TL;DR: A transfer learning framework based on sparse coding for pedestrian detection is proposed and it is shown that the trained scene-specific pedestrian detector performs well and is comparable with the detector trained on a large number of training samples manually labeled from the target scene.
Abstract: Pedestrian detection is a fundamental problem in video surveillance and has achieved great progress in recent years. However, training a generic detector performing well in a great variety of scenes has proved to be very difficult. On the other hand, exhausting manual labeling efforts for each specific scene to achieve high accuracy of detection is not acceptable especially for video surveillance applications. To alleviate the manual labeling efforts without scarifying accuracy of detection, we propose a transfer learning framework based on sparse coding for pedestrian detection. In our method, generic detector is used to get the initial target samples, and then several filters are used to select a small part of samples (called as target templates) from the initial target samples which we are very sure about their labels and confidence values. The relevancy between source samples and target templates and the relevancy between target samples and target templates are estimated by sparse coding and later used to calculate the weights for source samples and target samples. By adding the sparse coding-based weights to all these samples during re-training process, we can not only exclude outliers in the source samples, but also tackle the drift problem in the target samples, and thus get a well scene-specific pedestrian detector. Our experiments on two public datasets show that our trained scene-specific pedestrian detector performs well and is comparable with the detector trained on a large number of training samples manually labeled from the target scene.
11 citations
TL;DR: This paper proposes an structured dictionary learning algorithm to explicitly reveal the cluster structure of the dictionary and develops the SSPA algorithms with the structured dictionary besides non-structured one, and experiments show that the methods are efficient and outperform state-of-the-art graph-based SSL methods.
5 citations
TL;DR: Fitted spectral hashing is proposed, based on the fact that one-dimensional data of any distribution could be mapped to a uniform distribution without changing the local neighbor relations among data items, and could be fitted well by S-curve function and Fourier function.
5 citations
TL;DR: A two-step iterative representative selection algorithm, based on the assumption that the dataset can be approximately reconstructed by linear combinations of dictionary items, which can minimize this Kullback-Leibler (KL) divergence.
5 citations
01 Apr 2014
TL;DR: An adaptive background scale selection mechanism that simulates the background color distribution as the benchmark for color contrast and can extract more representative local regions with competitive repeatability score at only 50% computational time and 10% memory cost is proposed.
Abstract: In order to extract representative local invariant regions in textured natural images, we propose a Color-Contrast-MSER (CCM) detector with color-contrast pixel ranking, which can reduce the number of meaningless regions extracted from backgrounds. The main contributions are threefold: (1) In contrast with the original MSER[3] which adopts intensity pixel ranking, we develop a new pixel ranking mechanism based on color contrast analysis. (2) In this paper, the pixel ranking value of each pixel is defined as the color contrast between a kernel-sized window and the background. Therefore we propose an adaptive background scale selection mechanism that simulates the background color distribution as the benchmark for color contrast. (3) The experimental results demonstrate that compared with the original MSER detector[3], our Color-Contrast-MSER (CCM) detector can extract more representative local regions with competitive repeatability score at only 50% computational time and 10% memory cost.
1 citations