Search

Sign languages generally consist of a sequence of upper body gestures and are cooperative processes among various parts such as the hands, arms, and face. Therefore, the dynamics of the parts as well as the holistic appearance of the upper body and individual parts are essential for robust recognition. In this paper, a global–local representation (GLR) module is proposed to boost the spatiotemporal feature modeling. The GLR module is composed of global shift and local shift along the height, width, and temporal dimensions. Specifically, the global shift is applied to the entire feature map for holistic representation, while the local shift restricts itself to local patches to capture detailed features. Furthermore, a novel cross-scale aggregation module is designed to combine the global and local information in different dimensions. Extensive experimental results on three large-scale benchmarks, including WLASL, INCLUDE and LSA64, demonstrate that the proposed method achieves state-of-the-art recognition performance. [ABSTRACT FROM AUTHOR]

A novel metastable austenitic stainless steel denoted as 13Cr12Mn is obtained by adding 12 wt% Mn to 13Cr–5Ni–2Mo super‐martensitic stainless steel. The designed 13Cr12Mn exhibits a yield strength of 899 MPa, a tensile strength of 1068 MPa, and an elongation of 21.2%. The in situ hydrogen‐charging incremental step loading technique experiments are conducted to evaluate hydrogen embrittlement. The hydrogen‐induced cracking threshold stress of 13Cr12Mn is determined to be 961 MPa, which exceeds that of tempered S13Cr with a similar yield strength by 400 MPa. In addition, 13Cr12Mn also shows good resistance to general corrosion and pitting corrosion in alkaline solutions. [ABSTRACT FROM AUTHOR]

Analysis of human interaction is one important research topic of human motion analysis. It has been studied either using first-person vision (FPV) or third-person vision (TPV). However, the joint learning of both types of vision has so far attracted little attention. One of the reasons is the lack of suitable datasets that cover both FPV and TPV. In addition, existing benchmark datasets of either FPV or TPV have several limitations, including the limited number of samples, participant subjects, interaction categories, and modalities. In this work, we contribute a large-scale human interaction dataset, namely FT-HID dataset. FT-HID contains pair-aligned samples of first-person and third-person visions. The dataset was collected from 109 distinct subjects and has more than 90K samples for three modalities. The dataset has been validated by using several existing action recognition methods. In addition, we introduce a novel multi-view interaction mechanism for skeleton sequences, and a joint learning multi-stream framework for first-person and third-person visions. Both methods yield promising results on the FT-HID dataset. It is expected that the introduction of this vision-aligned large-scale dataset will promote the development of both FPV and TPV, and their joint learning techniques for human action analysis. [ABSTRACT FROM AUTHOR]

Quality checks are important to ensure the accuracy and reliability of spatial data. However, current methods primarily focus on attributes and geometric information, and methods to verify the accuracy of land-type classification for vector patches are lacking. Therefore, in this paper, we proposed a framework for the automatic verification of the land type of vector patches, including the segmentation of complex vector patches, automatic acquisition of training samples, and automatic extraction of suspicious patches based on deep learning scene interpretation results. First, an innovative method for segmentation based on an oriented bounding box was proposed. Then, scene interpretation from the segmented processing units was performed. Finally, the scene interpretation results and the original data for vector patches were combined, and segmentation units that do not match the category information were automatically identified. The suspicious patches were extracted, and the land-type authenticity of vector patches was automatically checked. Experimental results showed that, in the study area, the accuracy, precision, recall, and F1 values for the model based on first-level land categories were 0.989, 0.842, 0.906, and 0.873, respectively, and those for second-level land categories were 0.994, 0.770, 0.891, and 0.826, respectively. Accordingly, the newly developed method provides reliable technical support for checking the land-type authenticity of vector patches. [ABSTRACT FROM AUTHOR]

Acorus tatarinowii (A. tatarinowii) is a medicinal plant of the Araceae family. Currently, pharmacology focuses on the study of volatile oils, but there are few reports of another important secondary metabolite, lignan. Dirigent protein is thought to play an important role in plant secondary metabolism and responds to a variety of biotic and abiotic stresses. However, the DIR gene family of A. tatarinowii has not been systematically analyzed, and it is unknown whether it affects lignan synthesis. In this study, a total of 27 AtsDIRs were identified by comprehensive analysis of the genome of the medicinal plant A. tatarinowii , and the candidate gene AtsDIR2 3 that may be involved in lignan synthesis was screened through bioinformatics and transcriptome analysis. It is worth noting that AtsDIR23 is significantly expressed in rhizomes and is a member of the DIR-a subfamily. Subsequently, subcellular localization revealed that AtsDIR23 was localized in chloroplasts. The functional verification of AtsDIR2 3 b y the transient transformation of A. tatarinowii and the stable transformation of Arabidopsis thaliana showed that the content of lignans in overexpressed plants increased. Co-expression analysis screening revealed the MYB transcription factor (AtsMYB91) that is highly correlated with AtsDIR23 expression, while yeast one-hybrid assays and double luciferase experiments showed that AtsMYB91 negatively regulated the expression of AtsDIR2 3 b y binding to the AtsDIR23 promoter. In conclusion, AtsDIR23 can promote the accumulation of lignans, which provides a reference for further research on the regulation of lignans by DIR genes. [ABSTRACT FROM AUTHOR]

The effect of cryogenic plastic pre-strain (15 %, 30 %, and 45 % true strain) on the hydrogen embrittlement (HE) of an equimolar FeNiCoCrMn high-entropy alloy was studied through slow strain rate tension (SSRT) at 1 × 10−5 s−1. As the pre-strain increased, ductility loss under harsh electrochemical hydrogen charging showed a downward trend accompanied by the promotion of hydrogen adsorption. This work provided an effective strategy for achieving the synergy of high strength and improved HE resistance. The raw/processed data required to reproduce these findings cannot be shared at this time as the data is related to an ongoing study. [Display omitted] • Cryogenic pre-straining makes effective strengthening. • Hydrogen embrittlement is greatly alleviated in highly pre-strained FeNiCoCrMn. • The deformation twins act as stronger hydrogen traps than the dislocations. [ABSTRACT FROM AUTHOR]

In article number 2200582, Yu Yan, Yanjing Su, and co-workers design a novel Mn-added 13Cr austenitic stainless steel with high strength through a partial-recrystallization technique and test with in-situ hydrogen-charging incremental step loading technique, which increases hydrogen-induced cracking threshold stress by more than 30% while retains good general corrosion performance. [Extracted from the article]

Most current few-shot action recognition approaches follow the metric learning paradigm, measuring the distance of any sub-sequences (frames, any frame combinations or clips) between different actions for classification. However, this disordered distance metric between action sub-sequences ignores the long-term temporal relations of actions, which may result in significant metric deviations. What's more, the distance metric suffers from the distinctive temporal distribution of different actions, including intra-class temporal offsets and inter-class local similarity. In this paper, a novel few-shot action recognition framework, Frame-to-frame Temporal Alignment Network (FTAN), is proposed to address the above challenges. Specifically, an attention-based temporal alignment (ATA) module is devised to calculate the distance between corresponding frames of different actions along the temporal dimension to achieve frame-to-frame temporal alignment. Meanwhile, the Temporal Context module (TCM) is proposed to increase inter-class diversity by enriching the frame-level feature representation, and the Frames Cyclic Shift Module (FCSM) performs frame-level temporal cyclic shift to reduce intra-class inconsistency. In addition, we present temporal and global contrastive objectives to assist in learning discriminative and class-agnostic visual features. Experimental results show that the proposed architecture achieves state-of-the-art on HMDB51, UCF101, Something-Something V2 and Kinetics-100 datasets. • Frame-to-frame temporal alignment network for few-shot action recognition. • Multi-head self-attention captures inter-class local similarity. • Frames cyclic shift tackles intra-class temporal offsets. • Frame- and video-level similarity measure the distance between action sequences. • Temporal and global contrastive loss as auxiliary objectives. [ABSTRACT FROM AUTHOR]

BORON is a lightweight substitution–permutation network cipher proposed in 2017. We reduce the number of guessed key bits by key-bridging technology and first utilize Fast Walsh Transform on BORON to minimize the time complexity. Finally, this paper gives the better key-recovery attack against block cipher BORON than previously proposed by 2 rounds: we realize a 11-round key-recovery attack on BORON-80 and 13-round key-recovery attack on BORON-128. The attacks proposed in this paper are the best attacks against BORON-80/128 to date. [ABSTRACT FROM AUTHOR]

Vessel segmentation in optical coherence tomography angiography (OCTA) is crucial for the detection and diagnosis of various eye diseases. However, it is hard to distinguish intricate vessel morphology and quantify the density of blood vessels due to the large variety of vessel sizes, significant background noise, and small datasets. To this end, a retinal angiography multi-scale segmentation network, integrated with the inception and squeeze-and-excitation modules, is proposed to address the above challenges under the one-shot learning paradigm. Specifically, the inception module extends the receptive field and extracts multi-scale features effectively to handle diverse vessel sizes. Meanwhile, the squeeze-and-excitation module modifies channel weights adaptively to improve the vessel feature extraction ability in complex noise backgrounds. Furthermore, the one-shot learning paradigm is adapted to alleviate the problem of the limited number of images in existing retinal OCTA vascular datasets. Compared with the classic U2-Net, the proposed model gains improvements in the Dice coefficient, accuracy, precision, recall, and intersection over union by 3.74%, 4.72%, 8.62%, 4.87%, and 4.32% respectively. The experimental results demonstrate that the proposed one-shot learning method is an effective solution for retinal angiography image segmentation. [ABSTRACT FROM AUTHOR]

Demirci–Selçuk meet-in-the-middle (DS-MITM) attack is an effective method for cryptanalysis. As far as we know, the published automatic results of DS-MITM attack are all for byte-oriented ciphers. In this article, we first propose the automatic analysis method of DS-MITM attack for bit-oriented ciphers based on constraint programming, which is integrated with key-bridging technique. Based on the automatic modeling method, we propose the first result of DS-MITM attack on SIMON, which is a family of lightweight block ciphers proposed by the National Security Agency (NSA) in 2013. [ABSTRACT FROM AUTHOR]

A 13Cr austenitic stainless steel with a good combination of high yield strength, ductility and hydrogen embrittlement resistance was designed by combining hetero-deformation induced strengthening and martensitic transformation induced plasticity effects. Room temperature metastable austenite is promoted by adding 8 wt% Mn to a 13Cr–5Ni–2Mo supermartensitic stainless steel. A heterostructure consisting of remaining martensite grains located between lath austenite grains and micrometer-sized austenite grains embedded inside ultrafine grains, was fabricated by cold rolling (44% reduction) and annealing at 973 K through a shear reversion process. The novel 13Cr austenitic stainless steel exhibits a yield strength of 923 MPa, tensile strength of 1085 MPa and total elongation of 33.2%. The hydrogen embrittlement resistance of the novel steel was determined by incremental step loading technique testing and was increased 40% compared with that of the 13Cr–5Ni–2Mo supermartensitic stainless steel. This study provides an approach for the design and industrial fabrication of high strength (125 ksi grade level), high ductility and hydrogen embrittlement resistant stainless steels that can be used in harsh service environments containing hydrogen. [ABSTRACT FROM AUTHOR]

The low yield strength (~300 MPa) of twinning-induced plasticity (TWIP) steels greatly limits their structural applications in the industrial field. Conventional strengthening mechanisms usually cause an enhancement of yield strength but also a severe loss of ductility. In this research, gradient substructures were introduced in the Fe-22Mn-0.6C TWIP steels by different pre-torsional deformation in order to overcome the above limitations. The substructure evolution, mechanical properties, and their origins in gradient-substructured (GS) TWIP steels were measured and compared by electron backscattered diffraction (EBSD), monotonous and loading-unloading-reloading (LUR) tensile tests. It was found that a simple torsional treatment could prepare gradient twins and dislocations in coarse-grained TWIP steel samples depending on torsional strain. The uniaxial tensile tests indicated that a superior combination of high yield strength, high ultimate strength, and considerable ductility was simultaneously obtained in the GS samples. The high yield strength and high ultimate tensile strength were attributed to synergetic strengthening mechanisms, viz., dislocation strengthening, due to the accumulation of high density of dislocations, and very high back stress strengthening due to gradient substructure distribution, which was accommodated through pile-ups of extra geometrically necessary dislocations (GNDs) across the sample-scale. Additionally, high ductility originated from gradient substructure-induced back stress hardening. The present study is also beneficial to the design efforts of high strength and high ductility of other heterogeneous-structured TWIP alloy systems. [ABSTRACT FROM AUTHOR]

The effect of deformation twins on the hydrogen embrittlement (HE) of a twinning-induced plasticity (TWIP) steel with various pre-strain levels, coupled with subsequent annealing treatment was systematically investigated by a slow strain rate tension test after cathodic hydrogen charging. The results revealed that specimens with high twin volume fraction had relatively better HE resistance. The high-density twin boundary not only plays an important role in hindering hydrogen diffusion, but also can refine the grains and inhibit the accumulation of hydrogen, so as to relieve stress concentration and avoid premature fracture of the specimen. [Display omitted] • The larger the pre-strain, the higher the strength, and the better the HE resistance. • Sample annealing after high pre-strain still maintains good HE resistance. • Twins have an advantage over dislocations in hindering hydrogen diffusion. • High density of twin boundaries can disperse hydrogen to prevent strain localization. [ABSTRACT FROM AUTHOR]

This paper presents a study of automatic design of neural network achitectures for skeleton-based action recognition. Specifically, we encode a skeleton-based action instance into a tensor and carefully define a set of operations to build two types of network cells: normal cells and reduction cells. The recently developed DARTS (Differentiable Architecture Search) is adopted to search for an effective network architecture that is built upon the two types of cells. All operations are 2D based in order to reduce the overall computation and search space. Experiments on the challenging NTU RGB+D and Kinectics datasets have verified that most of the networks developed to date for skeleton-based action recognition are likely not compact and efficient. The proposed method provides an approach to search for such a compact network that is able to achieve comparative or even better performance than the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The origins of back stress strengthening were investigated in Fe-22Mn-0.6C and Fe-22Mn-0.6C-3Al TWIP steels. In contrast to reports found in literature, this study revealed that the high back stress developed in these steels is primarily caused by pile-ups of high density of geometrically necessary dislocations near the grain boundary regions rather than at deformation twins. • Back stress in 22Mn0.6C and 22Mn0.6C3Al TWIP steels was compared using LUR tests. • GND distributions across microstructure of steels were determined using EBSD data. • Deformation twins were less effective for GND accumulation than grain boundaries. • Deformation twins were not leading features responsible for high back stress. • High back stress originated from pile-ups of high GNDs near grain boundary regions. [ABSTRACT FROM AUTHOR]

Accurate extraction of buildings using high spatial resolution imagery is essential to a wide range of urban applications. However, it is difficult to extract semantic features from a variety of complex scenes (e.g., suburban, urban and urban village areas) because various complex man-made objects usually appear heterogeneous with large intra-class and low inter-class variations. The automatic extraction of buildings is thus extremely challenging. The fully convolutional neural networks (FCNs) developed in recent years have performed well in the extraction of urban man-made objects due to their ability to learn state-of-the-art features and to label pixels end-to-end. One of the most successful FCNs used in building extraction is U-net. However, the commonly used skip connection and feature fusion refinement modules in U-net often ignore the problem of feature selection, and the ability to extract smaller buildings and refine building boundaries needs to be improved. In this paper, we propose a trainable chain fully convolutional neural network (CFCN), which fuses high spatial resolution unmanned aerial vehicle (UAV) images and the digital surface model (DSM) for building extraction. Multilevel features are obtained from the fusion data, and an improved U-net is used for the coarse extraction of the building. To solve the problem of incomplete extraction of building boundaries, a U-net network is introduced by chain, which is used for the introduction of a coarse building boundary constraint, hole filling, and "speckle" removal. Typical areas such as suburban, urban, and urban villages were selected for building extraction experiments. The results show that the CFCN achieved recall of 98.67%, 98.62%, and 99.52% and intersection over union (IoU) of 96.23%, 96.43%, and 95.76% in suburban, urban, and urban village areas, respectively. Considering the IoU in conjunction with the CFCN and U-net resulted in improvements of 6.61%, 5.31%, and 6.45% in suburban, urban, and urban village areas, respectively. The proposed method can extract buildings with higher accuracy and with clearer and more complete boundaries. [ABSTRACT FROM AUTHOR]