GazeForensics: DeepFake detection via gaze-guided spatial inconsistency learning.

DeepFake detection is pivotal in personal privacy and public safety. With the iterative advancement of DeepFake techniques, high-quality forged videos and images are becoming increasingly deceptive. Prior research has seen numerous attempts by scholars to incorporate biometric features into the field of DeepFake detection. However, traditional biometric-based approaches tend to segregate biometric features from general ones and freeze the biometric feature extractor. These approaches resulted in the exclusion of valuable general features, potentially leading to a performance decline and, consequently, a failure to fully exploit the potential of biometric information in assisting DeepFake detection. Moreover, insufficient attention has been dedicated to scrutinizing gaze authenticity within the realm of DeepFake detection in recent years. In this paper, we introduce GazeForensics, an innovative DeepFake detection method that utilizes gaze representation obtained from a 3D gaze estimation model to regularize the corresponding representation within our DeepFake detection model, while concurrently integrating general features to further enhance the performance of our model. Experimental results demonstrate that our proposed GazeForensics method performs admirably in terms of performance and exhibits excellent interpretability.

Epigenetic control of skeletal muscle atrophy.

Skeletal muscular atrophy is a complex disease involving a large number of gene expression regulatory networks and various biological processes. Despite extensive research on this topic, its underlying mechanisms remain elusive, and effective therapeutic approaches are yet to be established. Recent studies have shown that epigenetics play an important role in regulating skeletal muscle atrophy, influencing the expression of numerous genes associated with this condition through the addition or removal of certain chemical modifications at the molecular level. This review article comprehensively summarizes the different types of modifications to DNA, histones, RNA, and their known regulators. We also discuss how epigenetic modifications change during the process of skeletal muscle atrophy, the molecular mechanisms by which epigenetic regulatory proteins control skeletal muscle atrophy, and assess their translational potential. The role of epigenetics on muscle stem cells is also highlighted. In addition, we propose that alternative splicing interacts with epigenetic mechanisms to regulate skeletal muscle mass, offering a novel perspective that enhances our understanding of epigenetic inheritance's role and the regulatory network governing skeletal muscle atrophy. Collectively, advancements in the understanding of epigenetic mechanisms provide invaluable insights into the study of skeletal muscle atrophy. Moreover, this knowledge paves the way for identifying new avenues for the development of more effective therapeutic strategies and pharmaceutical interventions.

HiFiSketch: High Fidelity Face Photo-Sketch Synthesis and Manipulation.

With the rapid development of generative adversarial networks, face photo-sketch synthesis has achieved promising performance and playing an increasingly important role in law enforcement as well as entertainment. However, most of the existing methods only work under the condition of no interference, and lack of generalization ability in wild scenes. The fidelity of the images generated by the existing methods are insufficient, and the manipulation ability according to text description is unavailable. Directly applying existing text-based image manipulation methods on face photo-sketch scenario may lead to severe distortions due to the cross-domain challenges. Therefore, we propose a novel cross-domain face photo-sketch synthesis framework named HiFiSketch, a network that learns to adjust the weights of generators for high-fidelity synthesis and manipulation. It can realize the translation of images between the photo domain and the sketch domain, and modify results according to the text input in the meanwhile. We further propose a cross-domain loss function, which can effectively preserve facial details during face photo-sketch synthesis. Extensive experiments on four public face sketch datasets show the superiority of our method compared to existing methods. We further present text-based face photo-sketch manipulation and sequential face photo-sketch manipulation for the first time to demonstrate the effectiveness of our method on high fidelity face photo-sketch synthesis and manipulation.

Design of a Biaxial High-G Piezoresistive Accelerometer with a Tension-Compression Structure.

To meet the measurement needs of multidimensional high-g acceleration in fields such as weapon penetration, aerospace, and explosive shock, a biaxial piezoresistive accelerometer incorporating tension-compression is meticulously designed. This study begins by thoroughly examining the tension-compression measurement mechanism and designing the sensor's sensitive structure. A signal test circuit is developed to effectively mitigate cross-interference, taking into account the stress variation characteristics of the cantilever beam. Subsequently, the signal test circuit of anti-cross-interference is designed according to the stress variation characteristics of the cantilever beam. Next, the finite element method is applied to analyze the structure and obtain the performance indices of the range, vibration modes, and sensitivity of the sensor. Finally, the process flow and packaging scheme of the chip are analyzed. The results show that the sensor has a full range of 200,000 g, a sensitivity of 1.39 µV/g in the X direction and 1.42 µV/g in the Y direction, and natural frequencies of 509.8 kHz and 510.2 kHz in the X and Y directions, respectively.

Rheological and Mechanical Properties of Bentonite-Cement Paste Reinforced with Basalt Fibers.

Bentonite cement paste (BCP) is among the grouting materials used widely in large-pore grouting and karst cave treatment. The mechanical properties of bentonite cement paste (BCP) will be improved by additional basalt fibers (BF). In this study, the effects of basalt fiber (BF) contents and their lengths on the rheological and mechanical properties of bentonite cement paste (BCP) have been examined. Yield stress (YS), plastic viscosity (PV), unconfined compressive strength (UCS), and splitting tensile strength (STS) were used to evaluate the rheological and mechanical properties of basalt fiber-reinforced bentonite cement paste (BFBCP). Scanning electron microscopy (SEM) and energy-dispersive x-ray spectroscopy (EDS) characterize microstructure development. The results indicate that the Bingham model can provide the rheological behavior of basalt fibers and bentonite cement paste (BFBCP). The yield stress (YS) and plastic viscosity (PV) increase as the content and length of basalt fiber (BF) increase. The effect of fiber content on yield stress (YS) and plastic viscosity (PV) is greater than that of fiber length. The addition of basalt fiber (BF) enhanced the unconfined compressive strength (UCS) and splitting tensile strength (STS) of basalt fiber-reinforced bentonite cement paste (BFBCP) at the optimum basalt fiber (BF) content of 0.6%. The optimum basalt fiber (BF) content tends to increase as curing age increases. The basalt fiber length of 9 mm is the most effective for improving unconfined compressive strength (UCS) and splitting tensile strength (STS). The large increments in unconfined compressive strength (UCS) and splitting tensile strength (STS) were 19.17% and 28.21% for the basalt fiber-reinforced bentonite cement paste (BFBCP), with a basalt fiber length of 9 mm and content of 0.6%. Scanning electron microscopy (SEM) shows that the randomly distributed basalt fiber (BF) forms a spatial network structure in basalt fiber-reinforced bentonite cement paste (BFBCP), which composes a stress system under the action of cementation. Basalt fibers (BF) used in crack generation processes slow down the flow through bridging and occur in the substrate to improve the mechanical properties of basalt fiber-reinforced bentonite cement paste (BFBCP).

Circulating circular RNA hsa_circ_0023179 acts as a diagnostic biomarker for non-small-cell lung cancer detection.

BACKGROUND: Lung cancer, the most prevalent cancer-related death worldwide, still lacks the means for early diagnosis. Because of the unique properties of the loop that make it stable in body fluids, circular RNAs (circRNAs) as a biomarker becomes a possibility. This research purposed to explore whether hsa_circ_0023179 can be applied as a possible biomarker for the early diagnosis and prognosis of non-small cell lung cancer (NSCLC). METHODS: hsa_circ_0023179 was screened by high-throughput sequencing of three pairs of NSCLC tissues and their surrounding tissues. Agarose gel electrophoresis (AGE), Sanger sequencing, exonuclease digestion assay, and actinomycin D were used to affirm the molecular properties of circRNA. Precision determination was performed by placement at room temperature and multiple freeze-thawing test for methodological evaluation. The expression of hsa_circ_0023179 in tissues, serum, and cells was determined by quantitative real-time polymerase chain reaction (qRT-PCR) to establish the receiver operating characteristic (ROC) curve to assess the diagnostic efficacy of hsa_circ_0023179. RESULTS: hsa_circ_0023179 conforms to the basic properties of circRNA, and the detection method of hsa_circ_0023179 has good stability and repeatability. Its expression was connected to histological type, TNM stage, lymph node metastasis, and distal metastasis in NSCLC tissues, serum, and cells. Compared with traditional tumor markers with higher sensitivity and specificity. A combined diagnosis can significantly improve the diagnostic value. The decrease in postoperative expression level suggests some potential for dynamic monitoring. CONCLUSION: hsa_circ_0023179 might be a promising novel serum marker for the detection and prediction of NSCLC.

Overexpression of lncRNA A2M-AS1 inhibits cell growth and aggressiveness via regulating the miR-587/bone morphogenetic protein 3 axis in lung adenocarcinoma.

Lung adenocarcinoma (LUAD) is a malignant tumor that occurs in the lungs. Numerous reports have substantiated the participation of long non-coding RNAs (lncRNAs) in the tumorigenesis of LUAD. Previously, lncRNA alpha-2-macroglobulin antisense RNA 1 (A2M-AS1) was confirmed to be an important regulator in the biological processes of LUAD and dysregulation of A2M-AS1 was associated with non-small cell lung cancer (NSCLC) progression. However, the precise mechanism of A2M-AS1 in LUAD has not been elucidated. Therefore, our study was designed to investigate the detailed molecular mechanism of A2M-AS1 in LUAD. Herein, the expression of lncRNA A2M-AS1, microRNA (miRNA) miR-587, and bone morphogenetic protein 3 (BMP3) in LUAD cell lines and tissues were detected by real-time quantitative polymerase chain reaction (RT-qPCR) and western blotting. The viability, proliferation, migration and invasion of LUAD cells were tested by cell counting kit-8 (CCK-8), colony formation and Transwell assays. In vivo tumor growth was investigated by xenograft animal experiment. Interactions among A2M-AS1, miR-587 and BMP3 were measured by RNA pulldown and luciferase reporter assays. In this study, A2M-AS1 was downregulated in LUAD tissues and cells and related to poor prognosis in LUAD patients. A2M-AS1 overexpression suppressed LUAD cell proliferation, migration and invasion in vitro and inhibited tumor growth in vivo. Mechanistically, A2M-AS1 directly bound with miR-587 to promote BMP3 expression in LUAD cells. Low expression of BMP3 was found in LUAD tissues and cells and was closely correlated with poor prognosis in LUAD patients. BMP3 deficiency reserved the inhibitory influence of A2M-AS1 overexpression on LUAD cell behaviors. Overall, A2M-AS1 inhibits cell growth and aggressiveness via regulating the miR-587/BMP3 axis in LUAD.

Knockdown of Long Noncoding RNA LINC00240 Inhibits Esophageal Cancer Progression by Regulating miR-26a-5p.

Background: Esophageal cancer is the most prevalent digestive system tumor. Due to a lack of characteristic symptoms and early diagnosis, a confirmed esophageal cancer is typically detected at a progressively harmful stage. Therefore, it is critical to investigate the molecular mechanisms governing the formation and progression of esophageal cancer in order to identify new treatment targets for esophageal cancer early detection. Methods: We first screened the differentially expressed gene LINC00240 in the TCGA database. Multivariate analysis and Cox regression were performed, and a nomogram was constructed for internal validation. The correlation between LINC00240 and immune cells was analyzed using the TIMER database. The possible mechanism of action was explored through GSEA enrichment analysis. Then, in 43 esophageal cancer tissues, paracancour tissues, and cell lines, the LINC00240 expression was found. Transwell assays, CCK-8, and clone formation assays were utilized to assess the impact of LINC00240 on the metastasis of esophageal cancer cells. The binding activity of LINC00240 to downstream miRNAs was assessed using the luciferase reporter gene. Results: TCGA database showed that LINC00240 expression was increased in cancer tissues compared to adjacent tissues. The C-index of the nomogram is 0.712 (0.666-0.758), and the prediction model has good accuracy. According to the TIMER database, the LINC00240 expression is linked to immune infiltration and may be crucial in encouraging the immune escape of tumor cells. Gene enrichment analysis depicts that LINC00240 could influence the biological events of esophageal cancer by taking part in pathways such as affecting the cell cycle. LINC00240 expression was substantially greater in the plasma of esophageal cancer patients (3.94 ± 1.55) than in the normal control group (2.13 ± 0.89). Plasma expression of LINC00240 was linked to the degree of differentiation (P=0.0345) and TNM stage (P=0.0409). Knocked down LINC00240 inhibited esophageal cancer cells proliferation, lone formation, and invasion. LINC00240 might bind itself to miR-26a-5p and influence its expression. MiR-26a-5p inhibitor can dramatically limit the ability of LINC00240 knockdown on plate colony formation and relocation of esophageal cancerous cells was demonstrated in colony formation and migration experiments. Conclusion: LINC00240 expression is elevated in esophageal cancerous tissues, and knocking down LINC00240 decreases esophageal cancer cell proliferation, clone formation, invasion, and migration via miR-26a-5p. As a result, LINC00240 could be a novel target for esophageal cancer patients' early diagnosis and treatment.

Heterogeneous Face Interpretable Disentangled Representation for Joint Face Recognition and Synthesis.

Heterogeneous faces are acquired with different sensors, which are closer to real-world scenarios and play an important role in the biometric security field. However, heterogeneous face analysis is still a challenging problem due to the large discrepancy between different modalities. Recent works either focus on designing a novel loss function or network architecture to directly extract modality-invariant features or synthesizing the same modality faces initially to decrease the modality gap. Yet, the former always lacks explicit interpretability, and the latter strategy inherently brings in synthesis bias. In this article, we explore to learn the plain interpretable representation for complex heterogeneous faces and simultaneously perform face recognition and synthesis tasks. We propose the heterogeneous face interpretable disentangled representation (HFIDR) that could explicitly interpret dimensions of face representation rather than simple mapping. Benefited from the interpretable structure, we further could extract latent identity information for cross-modality recognition and convert the modality factor to synthesize cross-modality faces. Moreover, we propose a multimodality heterogeneous face interpretable disentangled representation (M-HFIDR) to extend the basic approach suitable for the multimodality face recognition and synthesis. To evaluate the ability of generalization, we construct a novel large-scale face sketch data set. Experimental results on multiple heterogeneous face databases demonstrate the effectiveness of the proposed method.

The expression of heat shock protein A12B (HSPA12B) in non-Hodgkin's lymphomas.

BACKGROUND: Heat shock protein A12B (HSPA12B) plays a considerable protective role for cells, tissues, and organs against various noxious conditions. However, the expression of HSPA12B in cancer biology remains controversial. This study aimed to investigate the expression of HSPA12B and its role in cell adhesion mediated drug resistance (CAM-DR) of non-Hodgkin's lymphoma (NHL). METHODS: In this study, the expression of HSPA12B in NHL was determined by immunohistochemical, and the effect of HSPA12B expression on the prognosis of NHL was analyzed by Kaplan-Meier curves. Then, the transfection technique was used to research the effect of HSPA12B in cell apoptosis. The most important was to study the expression changes of HSPA12B in the adhesion model and the effect of overexpression of HSPA12B on CAM-DR. RESULTS: We analyzed the relationship between the expression levels of HSPA12B and clinical parameters in NHL. The expression of HSPA12B was directly related to the different NHL variants. We overexpressed HSPA12B in 2 NHL cell lines and found a subsequent reduction in apoptosis. More specifically, we used an adhesion assay to demonstrate that HSPA12B expression was induced in NHL cells when they adhered to fibronectin (FN) or bone marrow stroma cells (BMSCs). Finally, it was revealed that HSPA12B overexpression enhances CAM-DR. CONCLUSIONS: Our data suggest that HSPA12B may play a functional role in CAM-DR and is thus a potential novel target for NHL treatment.

Elucidating the Role of Serum tRF-31-U5YKFN8DYDZDD as a Novel Diagnostic Biomarker in Gastric Cancer (GC).

BACKGROUND: Gastric cancer (GC) is one of the malignant tumors with the highest morbidity and mortality in the world. Early diagnosis combined with surgical treatment can significantly improve the prognosis of patients. Therefore, it is urgent to seek higher sensitivity and specificity biomarkers in GC. tRNA-derived small RNAs are a new non-coding small RNA that widely exists in tumor cells and body fluids. In this study, we explore the expression and biological significance of tRNA-derived small RNAs in GC. MATERIALS AND METHODS: First of all, we screened the differentially expressed tRNA-derived small RNAs in tumor tissues by high-throughput sequencing. Agarose gel electrophoresis (AGE), Sanger sequencing, and Nuclear and Cytoplasmic RNA Separation Assay were used to screen tRF-31-U5YKFN8DYDZDD as a potential tumor biomarker for the diagnosis of GC. Then, we detected the different expressions of tRF-31-U5YKFN8DYDZDD in 24 pairs of GC and paracancerous tissues, the serum of 111 GC patients at first diagnosis, 89 normal subjects, 48 superficial gastritis patients, and 28 postoperative GC patients by quantitative real-time PCR (qRT-PCR). Finally, we used the receiver operating characteristic (ROC) curve to analyze its diagnostic efficacy. RESULTS: The expression of tRF-31-U5YKFN8DYDZDD has good stability and easy detection. tRF-31-U5YKFN8DYDZDD was highly expressed in tumor tissue, serum, and cell lines of GC, and the expression was significantly related to TNM stage, depth of tumor invasion, lymph node metastasis, and vascular invasion. The expression of serum tRF-31-U5YKFN8DYDZDD in the GC patients decreased after the operation (P = 0.0003). Combined with ROC curve analysis, tRF-31-U5YKFN8DYDZDD has better detection efficiency than conventional markers. CONCLUSIONS: The expressions of tRF-31-U5YKFN8DYDZDD in the tumor and paracancerous tissues, the serum of GC patients and healthy people, and the serum of GC patients before and after operation were different. tRF-31-U5YKFN8DYDZDD is not only a diagnostic biomarker of GC but also a predictor of poor prognosis.

LncRNA MAGI2-AS3 acts as a tumor suppressor that attenuates non-small cell lung cancer progression by targeting the miR-629-5p/TXNIP axis.

BACKGROUND: Lung cancer is a malignant tumor that seriously threatens the health of human beings. Long non-coding RNAs (lncRNAs) are thought to play important roles in the pathophysiology of lung cancer. In this study, we identified a new lncRNA, MAGI2-AS3 in non-small cell lung cancer (NSCLC) tissues by conducting an integrated bioinformatics analysis. Mechanistic studies were also performed to explore the biological functions of MAGI2-AS3 in NSCLC progression. METHODS: A bioinformatics analysis was conducted to determine the prognostic role of MAGI2-AS3. CCK-8, EdU assay, colony formation and Transwell were performed to determine the effects of MAGI2-AS3 on the progression of NSCLC cells. A nude mice model was used to evaluate the effects of MAGI2-AS2 on the in vivo tumor growth of NSCLC. Luciferase reporter and RNA pull-down assays were used to evaluate interactions between MAGI2-AS3 and its downstream targets. RESULTS: MAGI2-AS3 was found to be downregulated in NSCLC tissues. The gain-of-function in vitro studies showed that the overexpression of MAGI2-AS3 suppressed NSCLC cell proliferation and invasion. Conversely, the knockdown of MAGI2-AS3 had the opposite effects. The bioinformatics analysis and luciferase report assay revealed that MAGI2-AS3 functioned as competing endogenous RNA to suppress microRNA (miR)-629-5p expression, while miR-629-5p suppressed thioredoxin-interacting protein (TXNIP) expression by targeting its 3' untranslated region. The rescue experiment results showed that MAGI2-AS3 knockdown enhanced NSCLC cell progression (increasing cell proliferation and invasion, but reducing cell apoptosis), which was counteracted by miR-629-5p inhibition or TXNIP overexpression. CONCLUSIONS: The study revealed that MAGI2-AS3 was downregulated in NSCLC tissues and cells, and MAGI2-AS3 suppressed NSCLC cell progression. Further, the mechanistic results showed that MAGI2-AS3 exerted a tumor-suppressive effect in NSCLC by targeting the miR-629-5p/TXNIP axis.

Comprehensive Assessment of Serum hsa_circ_0070354 as a Novel Diagnostic and Predictive Biomarker in Non-small Cell Lung Cancer.

Background: More and more studies have shown that circular RNAs (circRNAs) play an essential role in the occurrence and development of tumors. Hence, they can be used as biomarkers to assist in diagnosing tumors. This study focuses on exploring the role of circular RNA (hsa_circ_0070354) in the diagnosis and prognosis of non-small cell lung cancer (NSCLC). Materials and Methods: First of all, high-throughput sequencing was used to find the difference in the expression of circular RNA between NSCLC and adjacent tissues. The circRNAs with higher differences in expression were selected to verify their expressions in tissues, cells, and serum using qRT-PCR. Secondly, the hsa_circ_0070354 with a significant difference was chosen as the research goal, and the molecular properties were verified by agarose gel electrophoresis and Sanger sequencing, etc. Then, actinomycin D and repeated freeze-thaw were used to explore the stability and repeatability of hsa_circ_0070354. Finally, the expression of hsa_circ_0070354 in serum of 133 patients with NSCLC and 97 normal donors was detected, and its sensitivity, specificity, and prognosis as tumor markers were statistically analyzed. Results: Hsa_circ_0070354 was highly expressed in tissues, cells, and serum of NSCLC, and it has the characteristics of sensitivity, stability, and repeatability. The ROC curve indicates that hsa_circ_0070354 is superior to conventional tumor markers in detecting NSCLC, and the combined diagnosis is of more significance in the diagnosis. The high expression of hsa_circ_0070354 is closely related to the late-stage, poor differentiation of the tumor and the short survival time of the patients, which is an independent indicator of poor prognosis. Conclusion: Hsa_circ_0070354 is not only a novel sensitive index for the diagnosis of NSCLC but also a crucial marker for bad biological behavior.

Hierarchically Porous SnO2 Coupled Organic Carbon for CO2 Electroreduction.

Converting CO2 into value-added chemicals or fuels by electrochemical CO2 reduction reaction (CO2 RR) has aroused great interest, whereas designing highly active and selective electrocatalysts is still a challenge. Herein, a novel kind of electrochemical catalyst composed with SnO2 and organic carbon (OC), named as SnO2 /OC, was facilely constructed for CO2 RR. The obtained SnO2 /OC exhibits both high faradaic efficiency for formate (∼75 %) and carbon products (∼95 %) as well as excellent stability. High surface area with hierarchically porous structure and the homogeneous formation of Sn-O-C linkages in SnO2 /OC jointly promote the adsorption and activation of CO2 , as well as fast transport of reactants and products.

Universal Face Photo-Sketch Style Transfer via Multiview Domain Translation.

Face photo-sketch style transfer aims to convert a representation of a face from the photo (or sketch) domain to the sketch (respectively, photo) domain while preserving the character of the subject. It has wide-ranging applications in law enforcement, forensic investigation and digital entertainment. However, conventional face photo-sketch synthesis methods usually require training images from both the source domain and the target domain, and are limited in that they cannot be applied to universal conditions where collecting training images in the source domain that match the style of the test image is unpractical. This problem entails two major challenges: 1) designing an effective and robust domain translation model for the universal situation in which images of the source domain needed for training are unavailable, and 2) preserving the facial character while performing a transfer to the style of an entire image collection in the target domain. To this end, we present a novel universal face photo-sketch style transfer method that does not need any image from the source domain for training. The regression relationship between an input test image and the entire training image collection in the target domain is inferred via a deep domain translation framework, in which a domain-wise adaption term and a local consistency adaption term are developed. To improve the robustness of the style transfer process, we propose a multiview domain translation method that flexibly leverages a convolutional neural network representation with hand-crafted features in an optimal way. Qualitative and quantitative comparisons are provided for universal unconstrained conditions of unavailable training images from the source domain, demonstrating the effectiveness and superiority of our method for universal face photo-sketch style transfer.

Coupled Attribute Learning for Heterogeneous Face Recognition.

Heterogeneous face recognition (HFR) is a challenging problem in face recognition and subject to large textural and spatial structure differences of face images. Different from conventional face recognition in homogeneous environments, there exist many face images taken from different sources (including different sensors or different mechanisms) in reality. In addition, limited training samples of cross-modality pairs make HFR more challenging due to the complex generation procedure of these images. Despite the great progress that has been achieved in recent years, existing works mainly focus on HFR from only cross-modality image matching. However, it is more practical to obtain both facial images and semantic descriptions about facial attributes in real-world situations, in which the semantic description clues are nearly always obtained during the process of image generation. Motivated by human cognitive mechanisms, we naturally utilize the explicit invariant semantic description, i.e., face attributes, to help address the gap among face images of different modalities. Existing facial attributes-related face recognition methods primarily regard attributes as the high-level features used to enhance recognition performance, ignoring the inherent relationship between face attributes and identities. In this article, we propose novel coupled attribute learning for the HFR (CAL-HFR) method without labeling the attributes manually. Deep convolutional networks are employed to directly map face images in heterogeneous scenarios to a compact common space where distances are taken as dissimilarities of pairs. Coupled attribute guided triplet loss (CAGTL) is designed to train an end-to-end HFR network that can effectively eliminate defects of incorrectly estimated attributes. Extensive experiments on multiple heterogeneous scenarios demonstrate that the proposed method achieves superior performance compared with that of state-of-the-art methods. Furthermore, we make publicly available our generated pairwise annotated heterogeneous facial attribute database for evaluation and promoting related research.

Engineering Active Fe Sites on Nickel-Iron Layered Double Hydroxide through Component Segregation for Oxygen Evolution Reaction.

Nickel-iron layered double hydroxide (NiFe LDH) is a promising oxygen evolution reaction (OER) electrocatalyst under alkaline conditions. Much research has been performed to understand the structure-activity relationship of NiFe LDH under OER conditions. However, the specific role of the Fe species remains unclear and under debate. Herein, based on DFT calculations, it was discovered that the edge Fe sites show higher activity towards OER than either the edge Ni sites or lattice sites. Therefore, a facile acid-etching method was proposed to controllably induce the formation of edge Fe sites in NiFe LDH, and the obtained sample exhibited higher OER activity. X-ray absorption near edge structure and extended X-ray absorption fine structure analyses further revealed that the interaction of the edge Fe species with Ni is believed to contribute to the enhancement of the OER performance. This work provides a new understanding of the structure-activity relationship in NiFe LDH and offers a facile method for the design of efficient electrocatalysts in an alkaline environment.

Re-ranking High-Dimensional Deep Local Representation for NIR-VIS Face Recognition.

Heterogeneous face recognition refers to matching facial images captured from different sensors or sources, which has wide applications in public security and law enforcement. Because of the great differences in sensing and creating procedure, there are huge feature gap between heterogeneous facial images. Existing methods merely focus on comparing the probe image with the gallery in feature space, while the true target may not appear at the first rank due to the appearance variations caused by different sensing patterns. In order to exploit valuable information from initial ranking result, this paper proposes to re-rank high-dimensional deep local representation for matching near-infrared (NIR) and visual (VIS) facial images, i.e. NIR-VIS face recognition. A high-dimensional deep local representation is firstly constructed by extracting and concatenating deep features on local facial patches via a convolutional neural network (CNN). The initial NIR-VIS recognition ranking results can be obtained by comparing the compressed deep features. We then propose a novel and efficient locally linear re-ranking (LLRe-Rank) technique to refine the initial ranking results, which can explore valuable information from initial ranking result. The proposed re-ranking method does not require any human interaction or data annotation, and can be served as an unsupervised post processing technique. Experimental results on the most challenging Oulu-CASIA NIR-VIS database and CASIA NIR-VIS 2.0 database demonstrate the effectiveness of our method.

Pt/Fe co-loaded mesoporous zeolite beta for CO oxidation with high catalytic activity and water resistance.

A series of Pt/Fe co-loaded mesoporous zeolite beta (Pt/Fe-mBeta) catalysts with different Fe contents have been successfully synthesized by an ion exchange and subsequent ethylene glycol reduction method. The catalysts were characterized by XRD, N2 adsorption-desorption, TEM, SEM, XPS and H2-TPR. The optimized sample Pt/Fe(3)-mBeta shows high catalytic activity for CO oxidation under dry conditions, and the complete conversion temperature of CO is as low as 90 °C. More importantly, the sample Pt/Fe(3)-mBeta also shows excellent water resistance and good durability, which could meet the practical needs of exhaust purification of diesel vehicles. It is believed that the synergistic effect between varied-valence Pt/Fe species and the mesoporous zeolite support with high surface area and good water resistance jointly contribute to the excellent catalytic performance.

Key Single-Atom Electrocatalysis in Metal-Organic Framework (MOF)-Derived Bifunctional Catalysts.

Metal-organic framework (MOF)-derived materials have attracted increasing interest and show promising catalytic performances in many fields. Intensive efforts have been focused on the structure design and metal-site integration in MOF-derived catalysts. However, the key catalytic processes related with the metal sites in MOF-derived catalysts that dominate the electrocatalytic performance still remain obscure. Herein, we show a neglected but critical issue in the pyrolytic synthesis of MOF-derived catalysts: the coupled evolution of dual sites, that is, metallic sites and single-atom metal sites. The identification of active sites of single-atom sites from the visible particles has been elucidated through the combined X-ray spectroscopic, electron microscopic, and electrochemical studies. Interestingly, after a total removal of metallic cobalt sites, catalysts with purified single-atom metal sites show no faltering activity for either the oxygen reduction reaction (ORR) or hydrogen evolution reaction (HER), while significantly enhanced ORR selectivity is achieved; this reveals the dominant activity and selectivity contribution from single-atom electrocatalysis. The insight of the coupled evolution of dual sites and the as-demonstrated dual-site decoupling strategies open up a new routine for the design and synthesis of MOF-derived catalysts with the optimized single-atom electrocatalysis towards various electrochemical reactions.