Aiming at addressing the problem of interactive gesture recognition between lunar robot and astronaut, a novel gesture detection and recognition algorithm is proposed. In gesture detection stage, based on saliency det...Aiming at addressing the problem of interactive gesture recognition between lunar robot and astronaut, a novel gesture detection and recognition algorithm is proposed. In gesture detection stage, based on saliency detection via Graph-Based Manifold Ranking (GBMR) algorithm, the depth information of foreground is added to the calculation of superpixel. By increasing the weight of connectivity domains in graph theory model, the foreground boundary is highlighted and the impact of background is weakened. In gesture recognition stage, Pyramid Histogram of Oriented Gradient (PHOG) feature and Gabor amplitude also phase feature of image samples are extracted. To highlight the Gabor amplitude feature, we propose a novel feature calculation by fusing feature in different directions at the same scale. Because of the strong classification capability and not-easy-to-fit advantage of Adaboosting, this paper applies it as the classifier to realize gesture recognition. Experimental results show that the improved gesture detection algorithm can maintain the robustness to influences of complex environment. Based on multi-feature fusion, the error rate of gesture recognition remains at about 4.2%, and the recognition rate is around 95.8%.展开更多
Brain-derived neurotrophic factor (BDNF) plays a crucial role in human obesity. Yet, the neural circuitry supporting the BDNF-mediated control of energy homeostasis remains largely undefined. To map key regions that m...Brain-derived neurotrophic factor (BDNF) plays a crucial role in human obesity. Yet, the neural circuitry supporting the BDNF-mediated control of energy homeostasis remains largely undefined. To map key regions that might provide inputs to or receive inputs from the paraventricular nucleus (PVN) BDNF neurons, a key type of cells in regulating feeding and thermogenesis, we used rabies virus-based transsynaptic labeling and adeno-associated virus based anterograde tracing techniques to reveal their whole-brain distributions. We found that dozens of brain regions provide dense inputs to or receive dense inputs from PVN BDNF neurons, including several known weight control regions and several novel regions that might be functionally important for the BDNF-mediated regulation of energy homeostasis.Interestingly, several regions show very dense reciprocal connections with PVN BDNF neurons, including the lateral septum, the preoptic area, the ventromedial hypothalamic nucleus, the paraventricular thalamic nucleus, the zona incerta, the lateral parabrachial nucleus, the subiculum, the raphe magnus nucleus, and the raphe pallidus nucleus. These strong anatomical connections might be indicative of important functional connections. Therefore, we provide an outline of potential neural circuitry mediated by PVN BDNF neurons, which might be helpful to resolve the complex obesity network.展开更多
文摘Aiming at addressing the problem of interactive gesture recognition between lunar robot and astronaut, a novel gesture detection and recognition algorithm is proposed. In gesture detection stage, based on saliency detection via Graph-Based Manifold Ranking (GBMR) algorithm, the depth information of foreground is added to the calculation of superpixel. By increasing the weight of connectivity domains in graph theory model, the foreground boundary is highlighted and the impact of background is weakened. In gesture recognition stage, Pyramid Histogram of Oriented Gradient (PHOG) feature and Gabor amplitude also phase feature of image samples are extracted. To highlight the Gabor amplitude feature, we propose a novel feature calculation by fusing feature in different directions at the same scale. Because of the strong classification capability and not-easy-to-fit advantage of Adaboosting, this paper applies it as the classifier to realize gesture recognition. Experimental results show that the improved gesture detection algorithm can maintain the robustness to influences of complex environment. Based on multi-feature fusion, the error rate of gesture recognition remains at about 4.2%, and the recognition rate is around 95.8%.
基金funded by the Pudong New Area Science and Technology Development Fund,China (J. Yang,#PKJ2014-Y08)Shanghai University of Medicine & Health Sciences Seed Foundation Project,China (J. Yang,#SPF-18-20-14-006)+2 种基金ational Natural Science Foundation of China (W. Shen,#31771169 and #91857104)Thousand Young Talents Program of China (W. Shen)ShanghaiTech University start-up fund (W. Shen)
文摘Brain-derived neurotrophic factor (BDNF) plays a crucial role in human obesity. Yet, the neural circuitry supporting the BDNF-mediated control of energy homeostasis remains largely undefined. To map key regions that might provide inputs to or receive inputs from the paraventricular nucleus (PVN) BDNF neurons, a key type of cells in regulating feeding and thermogenesis, we used rabies virus-based transsynaptic labeling and adeno-associated virus based anterograde tracing techniques to reveal their whole-brain distributions. We found that dozens of brain regions provide dense inputs to or receive dense inputs from PVN BDNF neurons, including several known weight control regions and several novel regions that might be functionally important for the BDNF-mediated regulation of energy homeostasis.Interestingly, several regions show very dense reciprocal connections with PVN BDNF neurons, including the lateral septum, the preoptic area, the ventromedial hypothalamic nucleus, the paraventricular thalamic nucleus, the zona incerta, the lateral parabrachial nucleus, the subiculum, the raphe magnus nucleus, and the raphe pallidus nucleus. These strong anatomical connections might be indicative of important functional connections. Therefore, we provide an outline of potential neural circuitry mediated by PVN BDNF neurons, which might be helpful to resolve the complex obesity network.
