Sun Yat-sen University

About: Sun Yat-sen University is a education organization based out in Guangzhou, Guangdong, China. It is known for research contribution in the topics: Population & Cancer. The organization has 115149 authors who have published 113763 publications receiving 2286465 citations. The organization is also known as: Zhongshan University & SYSU.

Visual Saliency Detection Based on Multiscale Deep CNN Features

Abstract: Visual saliency is a fundamental problem in both cognitive and computational sciences, including computer vision. In this paper, we discover that a high-quality visual saliency model can be learned from multiscale features extracted using deep convolutional neural networks (CNNs), which have had many successes in visual recognition tasks. For learning such saliency models, we introduce a neural network architecture, which has fully connected layers on top of CNNs responsible for feature extraction at three different scales. The penultimate layer of our neural network has been confirmed to be a discriminative high-level feature vector for saliency detection, which we call deep contrast feature. To generate a more robust feature, we integrate handcrafted low-level features with our deep contrast feature. To promote further research and evaluation of visual saliency models, we also construct a new large database of 4447 challenging images and their pixelwise saliency annotations. Experimental results demonstrate that our proposed method is capable of achieving the state-of-the-art performance on all public benchmarks, improving the F-measure by 6.12% and 10%, respectively, on the DUT-OMRON data set and our new data set (HKU-IS), and lowering the mean absolute error by 9% and 35.3%, respectively, on these two data sets.

Radical Covalent Organic Frameworks: A General Strategy to Immobilize Open-Accessible Polyradicals for High-Performance Capacitive Energy Storage

Abstract: Ordered π-columns and open nanochannels found in covalent organic frameworks (COFs) could render them able to store electric energy. However, the synthetic difficulty in achieving redox-active skeletons has thus far restricted their potential for energy storage. A general strategy is presented for converting a conventional COF into an outstanding platform for energy storage through post-synthetic functionalization with organic radicals. The radical frameworks with openly accessible polyradicals immobilized on the pore walls undergo rapid and reversible redox reactions, leading to capacitive energy storage with high capacitance, high-rate kinetics, and robust cycle stability. The results suggest that channel-wall functional engineering with redox-active species will be a facile and versatile strategy to explore COFs for energy storage.

Look into Person: Joint Body Parsing & Pose Estimation Network and a New Benchmark

Abstract: Human parsing and pose estimation have recently received considerable interest due to their substantial application potentials. However, the existing datasets have limited numbers of images and annotations and lack a variety of human appearances and coverage of challenging cases in unconstrained environments. In this paper, we introduce a new benchmark named “Look into Person (LIP)” that provides a significant advancement in terms of scalability, diversity, and difficulty, which are crucial for future developments in human-centric analysis. This comprehensive dataset contains over 50,000 elaborately annotated images with 19 semantic part labels and 16 body joints, which are captured from a broad range of viewpoints, occlusions, and background complexities. Using these rich annotations, we perform detailed analyses of the leading human parsing and pose estimation approaches, thereby obtaining insights into the successes and failures of these methods. To further explore and take advantage of the semantic correlation of these two tasks, we propose a novel joint human parsing and pose estimation network to explore efficient context modeling, which can simultaneously predict parsing and pose with extremely high quality. Furthermore, we simplify the network to solve human parsing by exploring a novel self-supervised structure-sensitive learning approach, which imposes human pose structures into the parsing results without resorting to extra supervision. The datasets, code and models are available at http://www.sysu-hcp.net/lip/ .

Low and non-platinum electrocatalysts for PEMFCs: Current status, challenges and prospects

Abstract: Platinum-based nanomaterials are the most commonly adopted electrocatalysts for both anode and cathode reactions in polymer electrolyte membrane fuel cells (PEMFCs) fed with hydrogen or low molecular alcohols. However, the scarce world reserves of Pt and its high price increases the total cost of the system and thus limits the feasibility of PEMFCs. Based on this problem, for PEMFCs to have wide practical applications and become commercially viable, the challenging issue of the high catalyst cost resulting from the exclusive adoption of Pt or Pt-based catalysts should be addressed. One of the targets of the scientific community is to reduce the Pt loading in membrane electrode assemblies (MEAs) to ca. 150 μ g c m MEA − 2 , simultaneously maintaining high PEMFCs performances. The present paper aims at providing the state-of-the-art of low Pt and non-Pt electrocatalysts for: (a) H2-O2 PEMFCs, (b) Direct Methanol Fuel Cells (DMFCs) and (c) Direct Ethanol Fuel Cells (DEFCs). The detailed analysis of a big number of recent investigations has shown that the highest mass specific power density (MSPD) value obtained for H2-O2 PEMFCs has far exceeded the 2015 target ( 5 mW μ g Pt total - 1 ) set by the USA department of energy, while a several number of investigations reported values between 1 and 5 mW μgPt−1. However, the highest values measured under DMFCs and DEFCs working conditions are still relatively low and close to 0.15 and 0.05 mW μgPt−1 respectively. Moreover, the last years, promising results have been reported concerning the design, fabrication, characterization, and testing of novel non-Pt (Pt-free) anodes and cathodes for PEMFCs applications.