Robust 3D Hand Detection from a Single RGB-D Image in Unconstrained Environments.

Three-dimensional hand detection from a single RGB-D image is an important technology which supports many useful applications. Practically, it is challenging to robustly detect human hands in unconstrained environments because the RGB-D channels can be affected by many uncontrollable factors, such as light changes. To tackle this problem, we propose a 3D hand detection approach which improves the robustness and accuracy by adaptively fusing the complementary features extracted from the RGB-D channels. Using the fused RGB-D feature, the 2D bounding boxes of hands are detected first, and then the 3D locations along the z-axis are estimated through a cascaded network. Furthermore, we represent a challenging RGB-D hand detection dataset collected in unconstrained environments. Different from previous works which primarily rely on either the RGB or D channel, we adaptively fuse the RGB-D channels for hand detection. Specifically, evaluation results show that the D-channel is crucial for hand detection in unconstrained environments. Our RGB-D fusion-based approach significantly improves the hand detection accuracy from 69.1 to 74.1 comparing to one of the most state-of-the-art RGB-based hand detectors. The existing RGB- or D-based methods are unstable in unseen lighting conditions: in dark conditions, the accuracy of the RGB-based method significantly drops to 48.9, and in back-light conditions, the accuracy of the D-based method dramatically drops to 28.3. Compared with these methods, our RGB-D fusion based approach is much more robust without accuracy degrading, and our detection results are 62.5 and 65.9, respectively, in these two extreme lighting conditions for accuracy.

Synthesizing Depth Hand Images with GANs and Style Transfer for Hand Pose Estimation.

Hand pose estimation is a critical technology of computer vision and human-computer interaction. Deep-learning methods require a considerable amount of tagged data. Accordingly, numerous labeled training data are required. This paper aims to generate depth hand images. Given a ground-truth 3D hand pose, the developed method can generate depth hand images. To be specific, a ground truth can be 3D hand poses with the hand structure contained, while the synthesized image has an identical size to that of the training image and a similar visual appearance to the training set. The developed method, inspired by the progress in the generative adversarial network (GAN) and image-style transfer, helps model the latent statistical relationship between the ground-truth hand pose and the corresponding depth hand image. The images synthesized using the developed method are demonstrated to be feasible for enhancing performance. On public hand pose datasets (NYU, MSRA, ICVL), comprehensive experiments prove that the developed method outperforms the existing works.

Accurate Hand Detection from Single-Color Images by Reconstructing Hand Appearances.

Hand detection is a crucial pre-processing procedure for many human hand related computer vision tasks, such as hand pose estimation, hand gesture recognition, human activity analysis, and so on. However, reliably detecting multiple hands from cluttering scenes remains to be a challenging task because of complex appearance diversities of dexterous human hands (e.g., different hand shapes, skin colors, illuminations, orientations, and scales, etc.) in color images. To tackle this problem, an accurate hand detection method is proposed to reliably detect multiple hands from a single color image using a hybrid detection/reconstruction convolutional neural networks (CNN) framework, in which regions of hands are detected and appearances of hands are reconstructed in parallel by sharing features extracted from a region proposal layer, and the proposed model is trained in an end-to-end manner. Furthermore, it is observed that the generative adversarial network (GAN) could further boost the detection performance by generating more realistic hand appearances. The experimental results show that the proposed approach outperforms the state-of-the-art on public challenging hand detection benchmarks.

Effects of Human-Social Capital Congruence and Environmental Dynamism on Dynamic of Encouragement and Organizational Innovation in New Ventures.

Although human capital and social capital can provide knowledge and social network for organizations, existing studies are inadequate to explore how the interaction between the two types of capital shapes organizational behaviors or organizational outcomes. The present study investigates whether the linkage of human capital to social capital was compensatory or complementary, and how they impact organizational innovation in consideration of the dynamic of encouragement. Using data from more than 200 technological new ventures in China, we analyze the associations among all the parameters through bootstrapping and response surface methods. The findings suggest that organizational innovation is stronger when human and social capital are congruent and that the dynamic of encouragement fully mediates the relationship between capital congruence and organizational innovation performance. Furthermore, environmental dynamism positively moderates the relationship between capital congruence and the dynamic of the environment, that is, the relationship is stronger for new ventures in high rather than low dynamic environments. Finally, the theoretical and managerial implications of this study are discussed.

Controllable synthesis of mesoporous carbon microspheres with renewable water glass as a template for lithium-sulfur batteries.

Mesoporous carbon microspheres (MCMs) were prepared via a spray drying-assisted template method using resorcinol-formaldehyde as the carbon precursor and water glass as the template. The pore structure could be controlled by adjusting the hydrolysis time, hydrolysis temperature, concentration of the water glass and reactant ratio. Water glass could be recycled after use, making this strategy environmentally friendly and cost-effective. MCMs with three-dimensional interconnected networks, high surface area (852-1549 m2 g-1), large pore volume (1.7-2.1 cm3 g-1) and controllable pore diameter (3.8-15.1 nm) were constructed and have good electrical conductivity and a large volume for sulfur loading. The S/MCM composites with abundant residual nanochannels could not only benefit for the diffusion of electrolyte but also improve the utilization of sulfur and buffer the volume expansion of sulfur. The MCMs with relatively small mesopores manifest a high reversible capacity and rate performance owing to the strong confinement effect of polysulfides. MCM-1 delivered an initial capacity of 888.7 mA h g-1 under 0.5C with a capacity retention of 700.5 mA h g-1 after 100 cycles. The good electrochemical performance confirms that mesoporous carbon microspheres can be an excellent host material for sulfur cathodes.

Low-Dose-Rate Brachytherapy Versus Cryotherapy in Low- and Intermediate-Risk Prostate Cancer.

PURPOSE: Cryotherapy and brachytherapy are definitive local treatment options for low- to intermediate-risk prostate cancer. There are both prospective and retrospective data for brachytherapy, but the use of cryotherapy has been limited primarily to single-institution retrospective studies. Currently, no published evidence has compared low-dose-rate brachytherapy versus cryotherapy. METHODS AND MATERIALS: Institutional review board approval was obtained to conduct a retrospective chart review of consecutive patients treated at our institution from 1990 to 2012. For inclusion, patients must have received a prostate cancer diagnosis and have been considered to have low- to intermediate-risk disease according to the National Comprehensive Cancer Network criteria. All patients received brachytherapy or cryotherapy treatment. Disease specifics and failure details were collected for all patients. Failure was defined as prostate-specific antigen nadir +2 ng/mL. RESULTS: A total of 359 patients were analyzed. The groups comprised 50 low-risk cryotherapy (LRC), 92 intermediate-risk cryotherapy (IRC), 133 low-risk brachytherapy (LRB), and 84 intermediate-risk brachytherapy (IRB) patients. The median prostate-specific antigen follow-up periods were 85.6 months (LRC), 59.2 months (IRC), 74.9 months (LRB), and 59.8 months (IRB). The 5-year biochemical progression-free survival (bPFS) rate was 57.9% in the cryotherapy group versus 89.6% in the brachytherapy group (P<.0001). The 5-year bPFS rate was 70.0% (LRC), 51.4% (IRC), 89.4% (LRB), and 89.7% (IRB). The bPFS rate was significantly different between brachytherapy and cryotherapy for low- and intermediate-risk groups (P<.05). The mean nadir temperature reached for cryotherapy patients was -35°C (range, -96°C to -6°C). Cryotherapy used a median of 2 freeze-thaw cycles (range, 2-4 freeze-thaw cycles). CONCLUSIONS: Results from this study suggest that cryotherapy is inferior to brachytherapy for patients with low- to intermediate-risk prostate cancer. Patient selection criteria for consideration of cryotherapy and brachytherapy are similar in terms of anesthesia candidacy. Therefore, cryotherapy would not be recommended as a first-line local therapy for this particular patient subset.

[Lead expelling effect of carboxymethyl chitosan calcium in lead poisoned mice].

OBJECTIVE: To observe the effects of carboxymethyl chitosan calcium (CCC) on concentration of lead, calcium and zinc, and the liver antioxidative capacity in lead poisoned mice. METHODS: Mice were randomly divided into 7 groups, including normal group, calcium carbonate group, lead-model group, and three experimental groups treated with CCC in three different doses, and the CaNa2EDTA positive control group. The lead poisoned mice model was established by giving water contained with lead acetate. CCC was administrated to mice i.g. once a day. Thirty days later, mice were killed and the concentrations of lead, calcium and zinc in blood, liver, brain and femur were determined by atomic absorption spectrophotometer. Maleic dialdehyde (MDA), total antioxidative capacity (T-AOC), superoxide dismutase (SOD), and glutathione peroxidase (GSH-Px) activities in liver were measured by using assay kit. RESULTS: CCC significantly reduced the concentration of lead in blood, brain, liver and femur from about 1.56 microg/g, 13.38 microg/g, 16.15 microg/g, 1011.62 microg/g to about 0.50 microg/g, 5.57microg/g, 5.64 microg/g, 457.86 microg/g, and markedly increased the concentration of calcium in femur in lead poisoned mice. CCC had no significant side-effects on concentration of zinc in lead poisoned mice. The antioxidative profile was favorably changed as manifested by decreasing the level of MDA, increasing the activities of SOD, GSH-Px and T-AOC in livers of the in lead poisoned mice. CONCLUSION: CCC might significantly advance the excretion of lead, increase the concentration of calcium in femur and the antioxidative capacity in lead-loaded mice.

Modulation of liver L-γ-glutamyl-L-cysteinylglycine homeostasis by N-acetyl-glucosamine-thiazolidine-4(R)-carboxylic acid in mice.

The properties of modulating liver L-γ-glutamyl-L-cysteinylglycine (GSH) homeostasis by thiazolidine derivative N-acetyl-glucosamine-thiazolidine-4(R)-carboxylic acid (GlcNAcCys) and the underlying mechanisms were investigated in L-buthionine-[S,R]-sulfoximine (BSO)-induced mice liver GSH depletion model. The data show that BSO (6 mmol/kg body weight; intraperitoneally) significantly decreased liver total sulfhydryl and GSH concentrations when compared with control. When mice were treated with different doses of GlcNAcCys (200, 400, 900 mg/kg body weight; intraperitoneally, respectively), total sulfhydryl and GSH concentrations were significantly increased when measured 6 hours after treatment. The activities of GSH-associated enzymes were also measured. Liver glutathione S-transferase (GST) activities were significantly decreased by BSO compared with the control, and GlcNAcCys significantly increased GST activity. Moreover, reverse-transcriptase polymerase chain reaction data indicated that GlcNAcCys could significantly induce glutamylcysteine ligase catalytic subunit c mRNA transcription. The mRNA levels of transcription factors c-jun and c-fos were increased by BSO administration but were decreased back to normal after the administration of GlcNAcCys. In a conclusion, GlcNAcCys can modulate liver GSH homeostasis, which may be related to its ability to induce glutamylcysteine ligase catalytic subunit transcription. GlcNAcCys has potential hepatoprotective properties by increasing GSH content, increasing GST activity.