Search

Existing monocular depth estimation methods have achieved satisfactory performance on wild datasets. However, these methods are usually trained and tested on a single dataset, which makes them difficult to generalize to other scenarios. To learn diverse scene priors from multiple datasets, we propose a hierarchical framework with adaptive bins for robust monocular depth estimation, which consists of two critical components: a group-wise query generator to assign hierarchical bins and a correlation-aware transformer decoder to generate adaptive bin features. The proposed HA-Bins enjoys several merits. First, the group-wise query generator progressively increases the number of bin queries for multi-scale image features, resulting in a hierarchical bin distribution robust to diverse scenarios. Second, the correlation-aware transformer decoder refines the correlation of bin queries and image features, effectively improving adaptive image feature aggregation. We visualize the query activation maps on NYUDepthv2 dataset, showing that the proposed network effectively suppresses the depth-irrelevant regions. Experiments on KITTI, Sintel, and RabbitAI benchmarks show that without any fine-tuning, our model jointly trained on multiple datasets achieves competitive performance with the state-of-the-art and solid robustness toward diverse scenarios. In addition, our method wins second place in Robust Vision Challenge 2022 towards challenging scenarios with different characteristics.

3D object detection in point cloud aims at simultaneously localizing and recognizing 3D objects from a 3D point set. However, since point clouds are usually sparse, unordered, and irregular, it is challenging to learn robust point representations and sample high-quality object queries. To deal with the above issues, we propose a Long-short rangE Adaptive transformer with Dynamic sampling (LeadNet), including a point representation encoder, a dynamic object query sampling decoder, and an object detection decoder in a unified architecture for 3D object detection. Specifically, in the point representation encoder, we combine an attention layer and a channel attentive kernel convolution layer to consider the local structure and the long-range context simultaneously. In the dynamic object query sampling decoder, we utilize multiple dynamic prototypes to adapt to various point clouds. In the object detection decoder, we incorporate a dynamic Gaussian weight map into the cross-attention mechanism to help the detection decoder focus on the proper visual regions near the object, further accelerating the training process. Extensive experimental results on two standard benchmarks show that our LeadNet outperforms the 3DETR baseline by $11.6\%~\text {mAP}_{50}$ on the ScanNet v2 dataset and achieves the new state-of-the-art results on ScanNet v2 and SUN RGB-D benchmarks for the geometric-only approaches.

Establishing effective correspondences between a pair of images is difficult due to real-world challenges such as illumination, viewpoint and scale variations. Modern detector-based methods typically learn fixed detectors from a given dataset, which is hard to extract repeatable and reliable keypoints for various images with extreme appearance changes and weakly textured scenes. To deal with this problem, we propose a novel Dynamic Keypoint Detection Network (DKDNet) for robust image matching via a dynamic keypoint feature learning module and a guided heatmap activator. The proposed DKDNet enjoys several merits. First, the proposed dynamic keypoint feature learning module can generate adaptive keypoint features via the attention mechanism, which is flexibly updated with the current input image and can capture keypoints with different patterns. Second, the guided heatmap activator can effectively fuse multi-group keypoint heatmaps by fully considering the importance of different feature channels, which can realize more robust keypoint detection. Extensive experimental results on four standard benchmarks demonstrate that our DKDNet outperforms state-of-the-art image-matching methods by a large margin. Specifically, our DKDNet can outperform the best image-matching method by 2.1% in AUC@ 3px on HPatches, 3.74% in AUC@$5^\circ$5∘ on ScanNet, 7.14% in AUC@$5^\circ$5∘ on MegaDepth and 12.32% in AUC@$5^\circ$5∘ on YFCC100M.

Visual tracking aims to estimate object state in a video sequence, which is challenging when facing drastic appearance changes. Most existing trackers conduct tracking with divided parts to handle appearance variations. However, these trackers commonly divide target objects into regular patches by a hand-designed splitting way, which is too coarse to align object parts well. Besides, a fixed part detector is difficult to partition targets with arbitrary categories and deformations. To address the above issues, we propose a novel adaptive part mining tracker (APMT) for robust tracking via a transformer architecture, including an object representation encoder, an adaptive part mining decoder, and an object state estimation decoder. The proposed APMT enjoys several merits. First, in the object representation encoder, object representation is learned by distinguishing target object from background regions. Second, in the adaptive part mining decoder, we introduce multiple part prototypes to adaptively capture target parts through cross-attention mechanisms for arbitrary categories and deformations. Third, in the object state estimation decoder, we propose two novel strategies to effectively handle appearance variations and distractors. Extensive experimental results demonstrate that our APMT achieves promising results with high FPS. Notably, our tracker is ranked the first place in the VOT-STb2022 challenge.

Modern information and communication systems have become increasingly challenging to manage. The ubiquitous system logs contain plentiful information and are thus widely exploited as an alternative source for system management. As log files usually encompass large amounts of raw data, manually analyzing them is laborious and error-prone. Consequently, many research endeavors have been devoted to automatic log analysis. However, these works typically expect structured input and struggle with the heterogeneous nature of raw system logs. Log parsing closes this gap by converting the unstructured system logs to structured records. Many parsers were proposed during the last decades to accommodate various log analysis applications. However, due to the ample solution space and lack of systematic evaluation, it is not easy for practitioners to find ready-made solutions that fit their needs. This paper aims to provide a comprehensive survey on log parsing. We begin with an exhaustive taxonomy of existing log parsers. Then we empirically analyze the critical performance and operational features for 17 open-source solutions both quantitatively and qualitatively, and whenever applicable discuss the merits of alternative approaches. We also elaborate on future challenges and discuss the relevant research directions. We envision this survey as a helpful resource for system administrators and domain experts to choose the most desirable open-source solution or implement new ones based on application-specific requirements.

In weakly supervised (WSAL) and unsupervised temporal action localization (UAL), the target is to simultaneously localize temporal boundaries and identify category labels of actions with only video-level category labels (WSAL) or category numbers in a dataset (UAL) during training. Among existing methods, attention based methods have achieved superior performance in both tasks by highlighting action segments with foreground attention weights. However, without the segment-level supervision on the attention weight learning, the quality of the attention weight hinders the performance of these methods. In this paper, we propose a novel Uncertainty Guided Collaborative Training (UGCT) strategy to alleviate this problem, which mainly includes two key designs: (1) The first design is an online pseudo label generation module, in which the RGB and FLOW streams work collaboratively to learn from each other. (2) The second design is an uncertainty aware learning module, which can mitigate the noise in the generated pseudo labels. These two designs work together to promote the model performance effectively and efficiently by exchanging information between RGB and FLOW streams. Extensive experimental results on two benchmark datasets with three attention based methods demonstrate the effectiveness of the proposed method, e.g, more than 7.0% performance gain for mAP@IoU=0.5 on THUMOS14 dataset.

Weakly supervised object localization (WSOL) aims to predict both object locations and categories with only image-level class labels. However, most existing methods rely on class-specific image regions for localization, resulting in incomplete object localization. To alleviate this problem, we propose a novel end-to-end task-aware framework with a transformer encoder-decoder architecture (TAFormer) to learn class-agnostic foreground maps, including a representation encoder, a localization decoder, and a classification decoder. The proposed TAFormer enjoys several merits. First, the designed three modules can effectively perform class-agnostic localization and classification in a task-aware manner, achieving remarkable performance for both tasks. Second, an optimal transport algorithm is proposed to provide pixel-level pseudo labels to online refine foreground maps. To the best of our knowledge, this is the first work by exploring a task-aware framework with a transformer architecture and an optimal transport algorithm to achieve accurate object localization for WSOL. Extensive experiments with four backbones on two standard benchmarks demonstrate that our TAFormer achieves favorable performance against state-of-the-art methods. Furthermore, we show that the proposed TAFormer provides higher robustness against adversarial attacks and noisy labels.

Quantitative molecular detection based on surface-enhanced Raman spectroscopy (SERS) is still a great challenge because of the highly nonuniform distribution of the SERS hot spots and the nondeterministic spatial and spectral overlap of the analyte with the hot spot. Here, we report a nanoparticle-on-mirror plasmonic sensor excited by surface plasmon polaritons for quantitative SERS detection of alpha fetoprotein in serum with ultrahigh sensitivity. The uniform gaps between the nanoparticles and gold film and the alignment of the gap modes relative to the excitation electric field endow this substrate with a uniform and strong SERS enhancement. The limit of detection reaches 1.45 fM, 697 times higher than that under normal excitation and 7800 times higher than a commercial enzyme-linked immunosorbent assay kit. This approach offers a potential solution to overcome the bottleneck in the field of SERS-based biosensing.

Common poly(lactide-co-glycolide) (i-PLGA) has emerged as a biodegradable and biocompatible material in tissue engineering. However, the poor hydrophilicity and elasticity of i-PLGA lead to its limited application in tissue engineering. To this end, an amphiphilic crosslinked four-armed poly(lactic-co-glycolide) was prepared. First, four-armed PLGA (4A-PLGA) was synthesized by polymerizing l-lactide (LA) and glycolide (GA) with pentaerythritol as the initiator. Then, the hydrophilic polymer poly(glutamate propylene ester) (PGPE) was prepared through the esterification of glutamic acid and 1,2-propanediol. The hydrophilic 4A-PLGA–PGPE was finally synthesized through the condensation reaction of 4A-PLGA and PGPE with the aid of triphosgene. 4A-PLGA–PGPE was then used to prepare amphiphilic membranes by electrospinning. It was demonstrated that the mechanical properties and biocompatibility of 4A-PLGA were improved after the introduction of PGPE. Furthermore, the introduction of glutamate improved the hydrophilicity of 4A-PLGA, thus effectively promoting cell entry and adhesion, which makes the electrospun 4A-PLGA–PGPE membranes promising for tissue engineering.

Prevention of Alzheimer’s disease (AD) is a global imperative, but reliable early interventions are currently lacking. Microglia-mediated chronic neuroinflammation is thought to occur in the early stage of AD and plays a critical role in AD pathogenesis. Here, oxytocin (OT)-loaded angiopep-2-modified chitosan nanogels (AOC NGs) were designed for early treatment of AD viainhibiting innate inflammatory response. Through the effective transcytosis of angiopep-2, AOC NGs were driven intravenously to cross the blood-brain barrier, enter the brain, and enrich in brain areas affected by AD. A large amount of OT was then released and specifically bound to the pathological upregulated OT receptor, thus effectively inhibiting microglial activation and reducing inflammatory cytokine levels through blocking the ERK/p38 MAPK and COX-2/iNOS NF-κB signaling pathways. Consecutive weekly intravenous administration of AOC NGs into 12-week-old young APP/PS1 mice, representing the early stage of AD, remarkably slowed the progression of Aβ deposition and neuronal apoptosis in the APP/PS1 mice as they aged and ultimately prevented cognitive impairment and delayed hippocampal atrophy. Together, the findings suggest that AOC NGs, which show good biosafety, can serve as a promising therapeutic candidate to combat neuroinflammation for early prevention of AD.

Hydrogel adhesives with integrated functionalities are still required to match their ever‐expanding practical applications in the field of tissue repair and regeneration. A simple and effective safety strategy is reported, involving an in situ injectable polymer precursor and visible light‐induced cross‐linking. This strategy enables the preparation of a hydrogel adhesive in a physiological environment, offering wet adhesion to tissue surfaces, molecular flexibility, biodegradability, biocompatibility, efficient hemostatic performance, and the ability to facilitate liver injury repair. The proposed one‐step preparation process of this polymer precursor involves the mixing of gelatin methacryloyl (GelMA), poly(thioctic acid) [P(TA)], poly(acrylic acid)/amorphous calcium phosphate (PAAc/ACP, PA) and FDA‐approved photoinitiator solution, and a subsequent visible light irradiation after in situ injection into target tissues that resulted in a chemically‐physically cross‐linked hybrid hydrogel adhesive. Such a combined strategy shows promise for medical scenarios, such as uncontrollable post‐traumatic bleeding. An injectable precursor solution containing a light initiator forms an adhesive hydrogel on tissues. Notably, the hydrogel possesses both biocompatibility and biodegradability, enabling safe and efficient prevention of tissue bleeding while promoting the repair of liver injuries.

To date, the robust and durable adhesion capability of hydrogel adhesives in wet environments remains a huge challenge. Herein, a physicochemically double-network crosslinked hydrogel matrix was prepared by mixing acrylic acid (AAc), chitosan (CS) and tannic acid (TA) as the main components and the subsequent in situpolymerization of AAc. The abundant reactive sites on the surface of the hydrogel matrix facilitate rapid, strong and repeatable adhesion to different surfaces of engineering solids and biological tissues in an aquatic environment. The formation of amide covalent bonds resulting from the addition of the bridging agent further expands the long-term application of the hydrogel in tissue repair, and the constructed hydrogel-tissue adhesive interface still has robust adhesion energy after soaking in a physiological environment for up to one month. Moreover, the hydrogel showed fantastic hemostatic performance due to its characteristics of platelet adhesion and high burst pressure. Overall, the persistent adhesion and excellent cytocompatibility of the hydrogel adhesive make it potentially applicable in medical adhesives.

A better understanding of cancer stem cells (CSCs) is essential for research on cancer therapy and drug resistance. Currently, increasingly more investigations are focused on obtaining CSCs to study the mechanism of their enhanced malignancy. In this work, three kinds of double-network hydrogels (PEMM/alginate), consisting of poly(ethylene glycol) (PEG) covalently cross-linked poly(methyl vinyl ether-alt-maleic acid) (P(MVE-alt-MA)) (network 1, denoted as PEMM) and Sr2+(or Ca2+, Fe3+) ionically cross-linked alginates (network 2, denoted as SrAlg, CaAlg, or FeAlg), were prepared. The stiffness, morphology, and components of the PEMM/alginate hydrogels were systematically investigated to understand their effects on CSC enrichment. Only the PEMM/FeAlg hydrogels could support the long-term growth, proliferation, and spheroid formation of SK-OV-3 cells. The expression of CSC-related markers was evaluated with the levels of protein and gene at different stages. The cell spheroids cultured in the PEMM/FeAlg hydrogels acquired certain CSC-like properties, thus drug resistance was enhanced, especially in the PEMM-1/FeAlg hydrogel. In vivotumorigenicity experiments also confirmed the presence of more CSCs in the PEMM-1/FeAlg hydrogel. The results suggest that matrix stiffness, morphology, and cations act synergistically on the regulation of the epithelial–mesenchymal transition (EMT), interleukin-6 (IL-6), and Wnt pathways, affecting the invasiveness of ovarian cancer and the conversion of the non-CSCs into CSCs. The PEMM-1/FeAlg hydrogel with lower elastic modulus, a more macroporous morphology, and higher swelling rate can significantly enhance the stemness, malignancy, and tumorigenicity of SK-OV-3 cells.

Polypropylene (PP) mesh has been used successfully for a long time in clinical practice as an impressive prosthesis for ventral hernia repair. To utilize a physical barrier for separating mesh from viscera is a general approach for preventing adhesions in clinical practice. However, a serious abdominal adhesion between the mesh and viscera can possibly occur post-hernia, especially with the small intestine; this can lead to a series of complications, such as chronic pain, intestinal obstruction, and fistula. Thus, determining how to prevent abdominal adhesions between the mesh and viscera is still an urgent clinical problem. In this study, a dopamine-functionalized polysaccharide derivative (oxidized-carboxymethylcellulose-g-dopamine, OCMC-DA) was synthesized; this was blended with carboxymethylchitosan (CMCS) to form a hydrogel (OCMC-DA/CMCS) in situ at the appropriate time. The physical and chemical properties of the hydrogel were characterized successfully, and its excellent biocompatibility was presented by the in vitro cell test. The combination of this hydrogel and PP mesh was used in laparoscopic surgery for repairing the abdominal wall defect, where the hydrogel could become fixed in situ on the PP mesh to form an anti-adhesion gel-mesh. The results showed that the gel-mesh could prevent abdominal adhesions effectively in the piglet model. Moreover, the histology and immunohistochemical staining proved that the gel-mesh could effectively alleviate the inflammation reaction and deposition of collagen around the mesh, and it did not disturb the integration between mesh and abdominal wall. Thus, the gel-mesh has superior tissue compatibility.

We report a superconducting nanowire single-photon detector (SNSPD) array aiming for a near-infrared 1550-nm wavelength that consists of 32 × 32 nanowire pixels and an area of 0.96 mm × 0.96 mm. Unlike most reported large-scale SNSPD arrays with amorphous films, NbN superconducting nanowires are employed in our array, which allows the detector operation at 2.3 K provided by a compact two-stage Gifford–McMahon cryocooler. Thermally coupled row–column multiplexing is employed in our arrays to avoid current redistribution and loss of electrical signal occurring in the electrically coupled row–column architecture. The fabricated detector array shows a pixel yield of 94% and maximal intrinsic efficiencies of 77% and 96% at 1550 nm and 405 nm, respectively. The timing jitter and the thermal coupling probability are also investigated.

We consider an integrated mobile gaming platform, in which the mobile device (e.g., smartphone) of a player can offload some game tasks toward a server as well as some neighboring mobile devices. The advantages of such a platform are manyfold: it can lead to an improved game experience, to a better use of energy resources, and, while offloading tasks to other mobile users, to the exploitation of the unused computing and storage resources of the mobile equipment, thus reducing the bandwidth and computing costs of the overall system. In this context, we formulate the problem of offloading the game computational tasks as an optimization problem that minimizes the maximum energy consumption across a set of mobile devices, under the constraints of a maximum response time and a limited availability of computation, communication, and storage resources. In light of the problem complexity, we then propose a heuristic, called Task Allocation with Minimal Energy cost (TAME), which is shown to closely approximate the optimal solution in all scenarios we considered. TAME also outperforms state-of-the-art algorithms under both synthetic and real scenarios, which have been devised based on a realistic and detailed energy consumption model for computation and communication resources. Our results, although tailored to mobile gaming, could be extended to other applications, where it may be beneficial to offload computational and storage tasks through device-to-device communications, as enabled by Wi-Fi, Bluetooth, or the upcoming 5G technology.

Polypropylene (PP) meshes are one of the most commonly used prosthesis materials in repairing abdominal wall defects. However, their application is usually limited due to possible serious abdominal adhesions between the mesh and the viscera. Instilling PP meshes with excellent anti-adhesion characteristics is still a formidable challenge. In this work, in order to prevent intestinal adhesion to the PP mesh, an effective method was developed to prepare anti-adhesive PP meshes, which was inspired by mussel adhesive proteins. A functional monomer, namely, dopamine methacrylamide, was first synthesized. Then, it was copolymerized with poly(ethylene glycol) methacrylate on the surface of O2-plasma-treated PP (OPP) meshes to form comb-like copolymer poly[poly(ethylene glycol) methacrylate-co-dopamine methacrylamide] (PEDMA), which was simultaneously grafted in situon the OPP mesh surface through the catechol group of PEDMA, subsequently yielding an anti-adhesive PP mesh (OPP-g-PEDMA). The properties of PEDMA and OPP-g-PEDMA meshes were characterized by NMR, GPC, TGA, FTIR, XPS, SEM, and water contact angle measurements. NIH-3T3 cells were employed to assess the cytocompatibility of OPP-g-PEDMA in vitro. Furthermore, the rat abdominal wall defect model was used to evaluate the efficacy of adhesion prevention. The results show that OPP-g-PEDMA not only possesses fantastic biocompatibility but also satisfactory anti-adhesion property involving minimal chronic inflammation, as well as lower adhesion formation rate and adhesion tenacity scores (less than 1.0). This type of OPP-g-PEDMA mesh is a promising candidate in effectively preventing peritoneal adhesion during abdominal wall defect repair.

Abdominal adhesion to polypropylene (PP) mesh remains one of the major complications in hernia repair. Thus, a challenge exists to endow PP mesh with powerful anti-adhesion properties in hernia repair. To investigate potential options, the assembled PP mesh was developed with effective anti-adhesion properties through an in situcoating of the mesh surface with self-fixable and biodegradable mussel-inspired hydrogels. Through mixing oxidized-carboxymethylcellulose functionalized with dopamine (OCMC-DA) with carboxymethylchitosan (CMCS), a layer of hydrogel (OCMC-DA/CMCS) can be formed in situon the PP mesh without the addition of crosslinking agents; the dopamine then acts as an immobilization group to fix these hydrogels to the PP mesh and the tissue surface. In this way, the assembled PP mesh (OCMC-DA/CMCS/PP) was obtained. The properties of the OCMC-DA/CMCS hydrogels were optimized, and the OCMC-DA4/CMCS hydrogel was selected to construct the assembled PP mesh. The lap-shear test revealed that OCMC-DA4/CMCS has tissue-adhesive properties. In vitrocell tests proved the excellent biocompatibility of the hydrogel. An optimized bioabsorption time and significant anti-adhesion properties were demonstrated through an in vivotest with a rat model. The adhesion area and tenacity of the OCMC-DA4/CMCS/PP group were more than 80% lower than those of the native PP mesh group and created a slightly inflammatory reaction.

Coherent multiwave mixing is in demand for optical frequency conversion, imaging, quantum information science, etc., but has rarely been demonstrated in solid-state systems. Here, we observed three- and five-wave mixing (5WM) in a c-axis growth zinc oxide microwire on a Au film with picosecond pulses in the near-infrared region. An output 5WM of 4.7 × 10–7μW, only 2–3 orders smaller than the three-wave mixing, is achieved when the excitation power is as low as 1.5 mW and the peak power density as weak as ∼107W/cm2. The excitation power dependence of 5WM agrees well with the perturbation limit under the low intensity but exhibits a strong deviation at a high pumping power. This extraordinary behavior is attributed to the cooperative resonant enhancement effect when pumping in the near-infrared range. Our study offers a potential solid-state platform for on-chip multiwave mixing and quantum nonlinear optics, such as generating many-photon entangled states or the construction of photon–photon quantum logic gates.

Cell-adhesive properties are of great significance to materials serving as extracellular matrix mimics. Appropriate cell-adhesive property of material interface can balance the cell–matrix interaction and cell–cell interaction and can promote cells to form 3D structures. Herein, a novel magnetic polyacrylamide (PAM) hydrogel fabricated via combining magnetostatic field induced magnetic nanoparticles assembly and hydrogel gelation was applied as a multicellular spheroids culturing platform. When cultured on the cell-adhesive microarray interface of sliced magnetic hydrogel, normal and tumor cells from different cell lines could rapidly form multicellular spheroids spontaneously. Furthermore, cells which could only form loose cell aggregates in a classic 3D cell culture model (such as hanging drop system) were able to be promoted to form multicellular spheroids on this platform. In the light of its simplicity in fabricating as well as its effectiveness in promoting formation of multicellular spheroids which was considered as a prevailing tool in the study of the microenvironmental regulation of tumor cell physiology and therapeutic problems, this composite material holds promise in anticancer drugs or hyperthermia therapy evaluation in vitro in the future.

China has become the largest producer of crude steel in the world since 1996, which places the country under huge pressure in terms of resources, energy, and the environment. Examining the driver of steel demand is of great significance to the structural adjustment and sustainable development of the steel industry. The researchers calculate the steel demand in China from 2000 to 2009 based on three sinks (steel stock, export, and loss) by taking the four stages of steel life cycle (production, fabrication and manufacturing, use, and waste management and recycling) as the study object. The researchers conclude that addition to in-use stock is the main driver of steel demand and that the 10-year average addition to inuse stock accounted for 77% of the steel sinks, in which 55% of the addition occurs in the building sector, and the steel for this segment is of low strength with large consumption. Based on the analysis of existing policies, the researchers propose that the steel demand structure will develop toward diversification and that the building sector will realize the upgrade of products as soon as possible to improve construction quality. Under the pressure of rising cost for imported resources, the export ratio of steel products should be controlled appropriately. Thus, recycling economy should be developed to reduce steel losses.

China has developed more than 20 water pollution control plans for river basins (RBWPs) since 1996. However, the implementation has generally lagged. This paper proposes a three-step, post-evaluation methodology to analyze the implementation result of a RBWP and its influential factors. First, a multi-attribute evaluation method based on an index system is established to score the enforcement results of a RBWP. Indicators measure how well a RBWP has achieved its objectives, which include water quality compliance, pollution load control, project construction, financial inputs, and related management requirements. Second, an interpretive structural model is used to detect the significant factors that affect RBWP implementation. This model can effectively analyze the cause-effect chain and hierarchical relationship among variables. Five groups of factors were identified, namely, plan preparation, water resource endowment, policy, institution, and management. Both qualitative and quantitative methods are employed in the third step to evaluate the extent to which these factors have influenced the execution result of a RBWP, including pre-post contrast, scenario analysis, and correlation analysis. This research then post-evaluated the implementation of the Huai River Basin water pollution control plans (HRBWPs) over a period of 10 years as a case study. Results showed that the implementation of the H-RBWPs was unsatisfactory during 2001–2005, although it improved during 2006–2010. The poor execution of these plans was partially caused by the underestimation of regional economic development in combination with ineffective industrial structure adjustment policies. Therefore, this case study demonstrates the feasibility and flexibility of the proposed post-evaluation methodology.

Vision-based sign language recognition has attracted more and more interest from researchers in the computer vision field. In this article, we propose a novel algorithm to model and recognize sign language performed in front of a Microsoft Kinect sensor. Under the assumption that some frames are expected to be both discriminative and representative in a sign language video, we first assign a binary latent variable to each frame in training videos for indicating its discriminative capability, then develop a latent support vector machine model to classify the signs, as well as localize the discriminative and representative frames in each video. In addition, we utilize the depth map together with the color image captured by the Kinect sensor to obtain a more effective and accurate feature to enhance the recognition accuracy. To evaluate our approach, we conducted experiments on both word-level sign language and sentence-level sign language. An American Sign Language dataset including approximately 2,000 word-level sign language phrases and 2,000 sentence-level sign language phrases was collected using the Kinect sensor, and each phrase contains color, depth, and skeleton information. Experiments on our dataset demonstrate the effectiveness of the proposed method for sign language recognition.

Crop residues are an important biomass, and are significant in the sustainable development of China. This paper uses the Grey-Markov modeling approach, the cost-benefit analysis method, and the constraint optimization method to establish the potential of crop residue recycling in China (CRRC) using a bottom-up analysis. Taking 2010 as the baseline year, the CRRC model is used to determine the quantity trends of crop residue resources, simulating the recycling potential and selecting key crop residue recycling technologies for operation between 2010 and 2030. The results illustrate that the total residue output from different crops will gradually increase to 1062 million tons in 2030. The proportion of crop residue for field burning is expected to decrease as a result of guidance and support from the government. Market mechanisms are also improving the development of the crop residue recycling industry. The economic benefit of crop residue recycling is expected to be worth 132 billion CNY in 2030 according to technology structure options. Key crop residue recycling technologies preferred such as liquefaction, amination, silo, co-firing straw power and composting will account for more than 85% of the total benefits.

For mimicking the fibrous extracellular matrix (ECM), a facile method for patterning anticell adhesive substrate was novelly applied on agarose hydrogel. Without using masks or templates for etching, we applied the magnetic field-induced colloidal assembly of magnetic nanoparticles on the flat agarose hydrogel to form cell-adhesive micropatterns. Meanwhile, tuning the hydrogel substrate’s modulus to fit real tissue was experimentally demonstrated. Magnetic nanobeads were also assembled on this hydrogel surface and formed more complete and regular patterns. The patterned hydrogel substrate could actually influence behaviors of different cancer cells, including adhesion, growth, and migration.

Given the statistical gaps in material flow among provinces in China, a method was introduced to estimate regional physical imports and exports (RPIE), which includes international and interregional imports/exports. This method uses provincial monetary inputoutput tables (MIOT) and international trade statistics. A coefficient matrix representing correlations between monetary value and physical mass for years 2000–2009 was obtained based on a detailed commodity classification and 22 material production sectors in MIOT. With the coefficient matrix as reference, RPIE was measured. Pilot calculation of both regional physical trade balance and domestic material consumption, as well as a brief analysis of these methods, were conducted using 2002 data.Given the statistical gaps in material flow among provinces in China, a method was introduced to estimate regional physical imports and exports (RPIE), which includes international and interregional imports/exports. This method uses provincial monetary inputoutput tables (MIOT) and international trade statistics. A coefficient matrix representing correlations between monetary value and physical mass for years 2000–2009 was obtained based on a detailed commodity classification and 22 material production sectors in MIOT. With the coefficient matrix as reference, RPIE was measured. Pilot calculation of both regional physical trade balance and domestic material consumption, as well as a brief analysis of these methods, were conducted using 2002 data.

To improve material efficiency, industrial structure optimization becomes a focal point in Chinese industrial and environmental policies. It is crucial to cluster economic sectors and determine their priority for industrial and environmental policy implementation. Integrating a set of criteria, a hybrid input-output model and the hierarchical cluster analysis, this study clusters China’s economic sectors and determines their priority on a life cycle basis. China’s economic sectors are clustered into three clusters. Industrial structure changes (industrial policy) should encourage the development of sectors in cluster 1 and limit the development of sectors in cluster 2. Technology development and materials recycling (two environmental policies) should mainly focus on sectors in clusters 1 and 2. Future industrial policies in China should limit the development of two sectors named Manufacture of metal products and Extraction of petroleum and natural gas. Instead of limiting some industries by command-and-control, the best policy option is to remedy environmental standards and law enforcement. Enterprises belonging to the identified key sectors from the viewpoint of direct production impacts should be concerned to achieve enterprise sustainability. To achieve sustainable production chains, the identified key sectors from the viewpoint of accumulative production impacts should be concerned. For sustainable consumption, the identified key sectors from the viewpoint of consumption impacts should be concerned to transform consumption styles. Most of environmental pressure can be alleviated not only by technical improvements and material recycling, but also by the development of economic sectors in cluster 1.To improve material efficiency, industrial structure optimization becomes a focal point in Chinese industrial and environmental policies. It is crucial to cluster economic sectors and determine their priority for industrial and environmental policy implementation. Integrating a set of criteria, a hybrid input-output model and the hierarchical cluster analysis, this study clusters China’s economic sectors and determines their priority on a life cycle basis. China’s economic sectors are clustered into three clusters. Industrial structure changes (industrial policy) should encourage the development of sectors in cluster 1 and limit the development of sectors in cluster 2. Technology development and materials recycling (two environmental policies) should mainly focus on sectors in clusters 1 and 2. Future industrial policies in China should limit the development of two sectors named Manufacture of metal products and Extraction of petroleum and natural gas. Instead of limiting some industries by command-and-control, the best policy option is to remedy environmental standards and law enforcement. Enterprises belonging to the identified key sectors from the viewpoint of direct production impacts should be concerned to achieve enterprise sustainability. To achieve sustainable production chains, the identified key sectors from the viewpoint of accumulative production impacts should be concerned. For sustainable consumption, the identified key sectors from the viewpoint of consumption impacts should be concerned to transform consumption styles. Most of environmental pressure can be alleviated not only by technical improvements and material recycling, but also by the development of economic sectors in cluster 1.

Based on our past work on the successfully fabricating giant liposomes, it is necessary to systematically study the capsules’ permeability. In this paper, fluorescence recovery after photo-bleaching (FRAP) experiment was carried out by confocal laser scanning microscope (CLSM) in order to quantify the 6-carboxyfluorescein (6-CF) molecules passing through capsules’ wall. The results showed that the diffusion coefficient of the pure multilayered polyelectrolyte (PE) capsules was similar to the one of partly the lipid-coated PE capsules. The diffusion coefficient for completely coated PE capsules catalyzed selectively by phospholipase A2is similar to that of partly coated PE capsules.

Summary: The colloidal stability of the aqueous dispersions of hydrophobic organic pigments, CuPc and carbon black, stabilized by a wide range of polymer structures based on alkyl vinyl ethers was studied. It was shown that, unlike the homopolymers and the random copolymers, the amphiphilic AB, ABA and BAB block copolymers of MVE with IBVE or ODVE show stabilizing activities that depend on their hydrophilic/hydrophobic balance and polymer architecture. After optimization, the colloidal stabilization is competitive with commercial PEO-PPO block copolymers (Pluronic®). It was found that the sedimentation of the dispersions was much faster at a higher temperature, above the LCST of the PMVE-blocks. The loss of the stabilizing activity of the block copolymers correlates with an increase of the hydrophobicity of the treated pigment surface. These properties enable the creation of colloidal dispersions with stabilities that can be tuned as a function of temperature.Poly(methyl vinyl ether) ABA and BAB block copolymers as colloidal stabilizers of organic pigments.

A means of correlation between the activity and the net charge, the metal atom net charge correlation (MANCC), which was successful in the activity prediction of the early‐transition metal catalysts, has been used to study the catalytic activities of salicylaldiminato Ni(II) complexes, the late‐transition metal catalysts, in olefin polymerization or oligomerization. A comparison with the available quantum mechanics/molecular mechanics (QM/MM) calculation data suggests that even without a detailed mechanism, MANCC results mostly agree with QM/MM calculation data regarding insertion barrier data and enthalpy change. Eight experimental complexes were further built up by modeling; their catalytic activities predominantly increased in line with the net charges on the metal atoms. The same results were obtained for the other four complexes synthesized in the present work. © 2004 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 42: 4765–4774, 2004

Four new phenoxy imino compounds were synthesized, and the solid-state structures of 1–3were determined by single-crystal X-ray diffraction study, giving crystallographic data as follows. 2-Acetyl-6-[1-(2,6-diisopropylphenylimino)ethyl]-4-methylphenol (1): a= 13.5893(5) Å, b= 10.5781(4) Å, c= 15.6778(4) Å, β = 113.1804(18)○, P21/c. 2,6-[1-(2,6-Diisopropylphenylimino)ethyl]-4-methylphenol (2): a= 12.1909(5) Å, b= 16.3324(6) Å, c= 15.9456(7) Å, β = 96.990(2)○, P21/c. 2-Acetyl-6-[1-(4-bromine-2,6-dimethylphenylimino)ethyl]-4-methylphenol (3): a= 7.5337(4) Å, b= 10.0457(5) Å, c= 12.6163(4) Å, α = 90.139(3)○, β = 104.003(3)○, γ = 106.485(2)○, P−1. Their molecular structures show that the 2,6-substituted phenyl ring is located approximately orthogonal to the hydroxyphenyl ring with the dihedral angle varying from 85.2○to 101.4○.

A combination of molecular mechanics and the charge equilibration method was applied and further developed to predicting the catalyst activity of a metal complex. A late transition metal catalyst, {di‐μ‐bromotetra [N,N′‐bis(3,5‐dimethylanil)‐4‐methyl‐2,6‐bis(imino)phenoxy]nickel} bromide (MOL) was dealt with. A modified molecular force field from universal force field was set and validated for this system. Simulations predict that the activity of MOL varies little with temperature. Simulation results were in good agreement with experimental results for ethylene oligomerization with MOL. Temperature dependence of QEq charge on the Ni central atom of MOL.

Poly(lactide-co-glycolide) (PLGA) has been widely used as a scaffold material for tissue engineering owing to its biocompatibility, biodegradability, and biosafety. However, lactic acid (LA) produced during PLGA degradation is prone to inflammation, which is a shortcoming that must be avoided. To this end, crystalline PLGA-PEG was synthesized here for the first time. To make the crystalline PLGA-PEG more suitable for tissue engineering, porous crystalline PLGA-PEG was prepared via the swelling behavior during recrystallization annealing. The structure and properties of the porous crystalline PLGA-PEG were confirmed by SEM, POM, and XRD. Furthermore, the swelling behavior of different PEG molecular weights was studied, and the cell viability test and alkaline phosphatase activity test showed that PLGA-PEG has good biocompatibility. Such a porous crystalline PLGA-PEG will make PLGA have a broader application prospect in bone repair.

Developing a high-performance nonprecious metal electrocatalyst for water splitting is a strong demand for the large-scale application of electrochemical H2production. In this work, we design a facile and scalable strategy to activate titanium metal for the hydrogen evolution reaction (HER) in alkaline media through incorporating hydrogen into the α-Ti crystal lattice by H2plasma bombardment. Benefiting from the accelerated charge transfer and enlarged electrochemical surface area after H2plasma treatment, the H-incorporated Ti shows remarkably enhanced HER activity with a much lower overpotential at −10 mA cm–2by 276 mV when compared to the pristine Ti. It is revealed that the retention of the incorporated H(D) atoms in the Ti crystal lattice during HER accounts for the durable feature of the catalyst. Density functional theory calculations demonstrate the effectiveness of hydrogen incorporation in tuning the adsorption energy of reaction species via charge redistribution. Our work offers a novel route to activate titanium or other metals by H incorporation through a controllable H2plasma treatment to tune the electronic structure for water splitting reactions.

Breast cancer, one of the most common diseases among women, is regarded as a heterogeneous and complicated disease that remains a major public health concern. Recently, owing to the development of next-generation sequencing technologies, long non-coding RNAs have received extensive attention. Numerous studies reveal that long non-coding RNAs are playing important roles in tumor development. Although the biological function and molecular mechanisms of long non-coding RNAs remain enigmatic, recent researchers have demonstrated that an array of long non-coding RNAs express abnormally in cancers, including breast cancer. Herein, we summarized the latest literature about long non-coding RNAs in breast cancer, with a particular focus on the multiple molecular roles of regulatory long non-coding RNAs that regulate cell proliferation, invasion, metastasis, and apoptosis.