GhHAM regulates GoPGF-dependent gland development and contributes to broad-spectrum pest resistance in cotton.

Cotton is a globally cultivated crop, producing 87% of the natural fiber used in the global textile industry. The pigment glands, unique to cotton and its relatives, serve as a defense structure against pests and pathogens. However, the molecular mechanism underlying gland formation and the specific role of pigment glands in cotton's pest defense are still not well understood. In this study, we cloned a gland-related transcription factor GhHAM and generated the GhHAM knockout mutant using CRISPR/Cas9. Phenotypic observations, transcriptome analysis, and promoter-binding experiments revealed that GhHAM binds to the promoter of GoPGF, regulating pigment gland formation in cotton's multiple organs via the GoPGF-GhJUB1 module. The knockout of GhHAM significantly reduced gossypol production and increased cotton's susceptibility to pests in the field. Feeding assays demonstrated that more than 80% of the cotton bollworm larvae preferred ghham over the wild type. Furthermore, the ghham mutants displayed shorter cell length and decreased gibberellins (GA) production in the stem. Exogenous application of GA3 restored stem cell elongation but not gland formation, thereby indicating that GhHAM controls gland morphogenesis independently of GA. Our study sheds light on the functional differentiation of HAM proteins among plant species, highlights the significant role of pigment glands in influencing pest feeding preference, and provides a theoretical basis for breeding pest-resistant cotton varieties to address the challenges posed by frequent outbreaks of pests.

Dietary antarctic krill improves antioxidant capacity, immunity and reduces lipid accumulation, insights from physiological and transcriptomic analysis of Plectropomus leopardus.

BACKGROUND: Due to its enormous biomass, Antarctic krill (Euphausia superba) plays a crucial role in the Antarctic Ocean ecosystem. In recent years, Antarctic krill has found extensive application in aquaculture, emerging as a sustainable source of aquafeed with ideal nutritional profiles. However, a comprehensive study focused on the detailed effects of dietary Antarctic krill on aquaculture animals, especially farmed marine fishes, is yet to be demonstrated. RESULTS: In this study, a comparative experiment was performed using juvenile P. leopardus, fed with diets supplemented with Antarctic krill (the krill group) or without Antarctic krill (the control group). Histological observation revealed that dietary Antarctic krill could reduce lipid accumulation in the liver while the intestine exhibited no obvious changes. Enzyme activity measurements demonstrated that dietary Antarctic krill had an inhibitory effect on oxidative stress in both the intestine and the liver. By comparative transcriptome analysis, a total of 1,597 and 1,161 differentially expressed genes (DEGs) were identified in the intestine and liver, respectively. Functional analysis of the DEGs showed multiple enriched terms significantly related to cholesterol metabolism, antioxidants, and immunity. Furthermore, the expression profiles of representative DEGs, such as dhcr7, apoa4, sc5d, and scarf1, were validated by qRT-PCR and fluorescence in situ hybridization. Finally, a comparative transcriptome analysis was performed to demonstrate the biased effects of dietary Antarctic krill and astaxanthin on the liver of P. leopardus. CONCLUSIONS: Our study demonstrated that dietary Antarctic krill could reduce lipid accumulation in the liver of P. leopardus, enhance antioxidant capacities in both the intestine and liver, and exhibit molecular-level improvements in lipid metabolism, immunity, and antioxidants. It will contribute to understanding the protective effects of Antarctic krill in P. leopardus and provide insights into aquaculture nutritional strategies.

Custom-Design of Multi-Stimuli-Responsive Degradable Silica Nanoparticles for Advanced Cancer-Specific Chemotherapy.

Chemotherapy is crucial in oncology for combating malignant tumors but often encounters obatacles such as severe adverse effects, drug resistance, and biocompatibility issues. The advantages of degradable silica nanoparticles in tumor diagnosis and treatment lie in their ability to target drug delivery, minimizing toxicity to normal tissues while enhancing therapeutic efficacy. Moreover, their responsiveness to both endogenous and exogenous stimuli opens up new possibilities for integrating multiple treatment modalities. This review scrutinizes the burgeoning utility of degradable silica nanoparticles in combination with chemotherapy and other treatment modalities. Commencing the elucidation of degradable silica synthesis and degradation mechanisms, emphasis is placed on the responsiveness of these materials to endogenous (e.g., pH, redox reactions, hypoxia, and enzymes) and exogenous stimuli (e.g., light and high-intensity focused ultrasound). Moreover, this exploration delves into strategies harnessing degradable silica nanoparticles in chemotherapy alone, coupled with radiotherapy, photothermal therapy, photodynamic therapy, gas therapy, immunotherapy, starvation therapy, and chemodynamic therapy, elucidating multimodal synergies. Concluding with an assessment of advances, challenges, and constraints in oncology, despite hurdles, future investigations are anticipated to augment the role of degradable silica in cancer therapy. These insights can serve as a compass for devising more efficacious combined tumor treatment strategies.

Severe hepatic encephalopathy with mechanical ventilation may inform waitlist priority in acute liver failure: A UNOS database analysis.

BACKGROUND & AIMS: Patients with acute liver failure (ALF) awaiting liver transplantation (LT) may develop multiorgan failure, but organ failure does not impact waitlist prioritization. The aim of this study was to examine the impact of organ failure on waitlist mortality risk and post LT outcomes in patients with ALF. METHODS: We studied adults waitlisted for ALF in the United Network for Organ Sharing (UNOS) database (2002-2019). Organ failures were defined using a previously described Chronic Liver Failure modified sequential organ failure score assessment adapted to UNOS data. Regression analyses of the primary endpoints, 30-day waitlist mortality (Competing risk), and post-LT mortality (Cox-proportional hazards), were performed. Latent class analysis (LCA) was used to determine the organ failures most closely associated with 30-day waitlist mortality. RESULTS: About 3212 adults with ALF were waitlisted, for hepatotoxicity (41%), viral (12%) and unspecified (36%) etiologies. The median number of organ failures was three (interquartile range 1-3). Having ≥3 organ failures (vs. ≤2) was associated with a sub hazard ratio (HR) of 2.7 (95%CI 2.2-3.4)) and a HR of 1.5 (95%CI 1.1-2.5)) for waitlist and post-LT mortality, respectively. LCA identified neurologic and respiratory failure as most impactful on 30-day waitlist mortality. The odds ratios for both organ failures (vs. neither) were higher for mortality 4.5 (95% CI 3.4-5.9) and lower for delisting for spontaneous survival .5 (95%CI .4-.7) and LT .6 (95%CI .5-.7). CONCLUSION: Cumulative organ failure, especially neurologic and respiratory failure, significantly impacts waitlist and post-LT mortality in patients with ALF and may inform risk-prioritized allocation of organs.

Scanning-based compressive hyperspectral imaging via spectral-coded illumination.

In this Letter, we present a novel, to the best of our knowledge, scanning-based compressive hyperspectral imaging method via spectral-coded illumination. We achieve efficient and flexible spectral modulation by spectral coding of a dispersive light source while spatial information is obtained by point-wise scanning, which can be applied to optical scanning imaging systems such as lidar. In addition, we propose a new tensor-based joint hyperspectral image reconstruction algorithm that considers spectral correlation and spatial self-similarity to recover three-dimensional hyperspectral data from compressive sampled data. Both simulated and real experiments show that our method has superior performance in visual quality and quantitative analysis.

Efficacy of vagus nerve stimulation in 95 children of drug-resistant epilepsy with structural etiology.

Vagus nerve stimulation (VNS) is one of the treatment options for drug-resistant epilepsy (DRE). To analyze the efficacy of VNS in children of DRE with structural etiology, we conducted a cohort study including 95 patients of DRE with structural etiology who underwent VNS treatment. Patients were followed up every 3 months at the outpatient department or via a remote programming platform. The median follow-up period was 2.6 years (range 1.0-4.6 years). The respective responder rates at 6, 12, 18, and 24 months of follow-up were 40.0% (38/95), 52.6% (50/95), 56.0% (47/84), and 59.7% (37/62). The respective seizure-free rates at 12, 18, and 24 months of follow-up were 8.4% (8/95), 9.5% (8/84), and 9.7% (6/62). The patients were divided into four groups based on etiologies: malformations of cortical development (n = 26), post-encephalitic lesions (n = 36), perinatal brain injury lesions (n = 31), and hippocampal sclerosis (n = 2). The respective responder rates at 12 months of follow-up in these groups were 53.8% (14/26), 52.8% (19/36), 51.6% (16/31), and 50.0% (1/2). There were no significant differences in gender, age at onset, age at stimulator implantation, epilepsy duration prior to VNS implantation, number of anti-seizure medications ever tried before VNS treatment, pulse amplitude of VNS, specific structural etiologies, lobe distribution or hemispheric side of structural lesions between responders and non-responders. Of the 95 patients, 8 (8.4%) underwent lesion surgery or hemispherectomy before VNS implantation, and 6/8 (75%) of these patients had a >50% reduction in seizure frequency. One patient who had a corpus callosotomy before VNS implantation had no response to VNS treatment. In conclusion, VNS is an effective treatment in children of DRE with structural etiology. There was no significant difference in VNS efficacy in patients with different structural etiologies. Vagus nerve stimulation treatment may also control seizures well in some patients with poor outcomes after lesion resection or hemispherectomy before VNS implantation.

The colonic metabolism differences of main alkaloids in normal and colitis mice treated with Coptis chinensis Franch. and Sophora flavescens Ait. herbal pair using liquid chromatography-high resolution mass spectrometry method combined with chemometrics.

Coptis chinensis Franch. and Sophora flavescens Ait. is a herbal pair frequently used in treating ulcerative colitis. However, the bio-disposition profile of the major components in the inflamed gut remains unclear, which is essential to understand the pharmacological material basis of this herb pair. Here we established an integral quantitative and chemometric method to deduce the colonic metabolism differences of this herbal pair in normal and colitis mice. With this LC-MS method, a total of 41 components have been found in the Coptis chinensis Franch. and Sophora flavescens Ait. extract, and 28 metabolites were found in the colon after oral administration. Alkaloid and its phase I metabolites were the main components in the colon of normal and colitis mice. The results of principal component analysis at 6 h after oral administration showed significant colonic metabolism differences between normal and colitis mice. Heamap results showed that colitis induced significant changes in the colonic bio-disposition of this herbal pair extract. In particular, in the context of colitis, the phase I metabolism of berberine, coptisine, jatrorrhizine, palmatine,and epiberberine has been inhibited. These results may provide a basis for understanding the pharmacological material basis of Coptis chinensis Franch. and Sophora flavescens Ait. in treating ulcerative colitis.

"RESET" Effect: Random Extending Sequences Enhance the Trans-Cleavage Activity of CRISPR/Cas12a.

The trans-cleavage activity of CRISPR/Cas12a has been widely used in biosensing applications. However, the lack of exploration on the fundamental properties of CRISPR/Cas12a not only discourages further in-depth studies of the CRISPR/Cas12a system but also limits the design space of CRISPR/Cas12a-based applications. Herein, a "RESET" effect (random extending sequences enhance trans-cleavage activity) is discovered for the activation of CRISPR/Cas12a trans-cleavage activity. That is, a single-stranded DNA, which is too short to work as the activator, can efficiently activate CRISPR/Cas12a after being extended a random sequence from its 3'-end, even when the random sequence folds into secondary structures. The finding of the "RESET" effect enriches the CRISPR/Cas12a-based sensing strategies. Based on this effect, two CRISPR/Cas12a-based biosensors are designed for the sensitive and specific detection of two biologically important enzymes.

Eight-Fold Increase in Dietary Supplement-Related Liver Failure Leading to Transplant Waitlisting Over the Last Quarter Century in the United States.

We investigated the trends in listing and outcomes of drug-induced acute liver failure (DIALF) over the last quarter century in the United States using the United Network for Organ Sharing (UNOS) database. We examined waitlisted patients in the UNOS database between 1995 and 2020 with a diagnosis of DIALF and assessed trends in etiologies, demographic and clinical characteristics, and outcomes over 3 periods: 1995-2003, 2004-2012, and 2013-2020. Patients with DIALF and cirrhosis were classified as drug-induced acute-on-chronic liver failure. Implicated agents including acetaminophen (APAP) and herbal or dietary supplements (HDSs) were ascertained. There were 2146 individuals with DIALF during the study period. The observed demographic trends between the earliest and latest period included fewer pediatric patients (18.8% to 13.5%) but with an increasing number of males in non-APAP DIALF (31.8% to 41.4%) and increased racial diversity in APAP DIALF. Antimicrobials remained the most common non-APAP agents across all periods, but antiepileptics, propylthiouracil, and mushroom poisoning decreased, while HDSs markedly increased from 2.9% to 24.1% of all non-APAP DIALF patients. The overall 5-year post-liver transplantation (LT) patient survival improved significantly over the 3 periods (69.9% to 77.4% to 83.3%) and was evident for both APAP and non-APAP DIALF. Over the last quarter century, there has been an 8-fold increase in HDS-related liver failure necessitating waitlisting for liver transplantation in the United States. There are other important temporal trends during the study period, including improved survival following LT among both APAP and non-APAP DIALF patients.

MoG-DS: model-guided deep convolutional network for joint denoising and super-resolution of a single-photon counting image.

Single-photon counting (SPC) imaging has attracted considerable research attention in recent years due to its capability to detect targets under extremely low-light conditions. However, the spatial quality of SPC images is always unsatisfactory because they typically suffer from considerable effects of noise and their spatial resolution is low. Most traditional methods are dedicated to solving the noise problem while ignoring the improvement of spatial resolution. To address these challenging issues, we propose a novel model-guided deep convolutional network for joint denoising and super-resolution (SR) of SPC images. First, we introduce a model-based iterative optimization algorithm with deep regularizer to unify denoising and SR into one problem. Second, we construct a model-guided deep convolutional network by unfolding the aforementioned model-based iterative algorithm to achieve an optimal solution. All modules in the proposed network are interpretable due to the special model-guided design, and they enable good generalization in real situations. In addition, the deep regularizer and other parameters in the proposed network are jointly optimized in an end-to-end manner, which efficiently reduces the difficulty of parameter design. Extensive simulation and real experimental results are reported to demonstrate the superiority of the proposed method in terms of visual comparison and quantitative analysis, respectively.

Line-wise scanning-based super-resolution imaging.

In this Letter, we present a novel, to the best of our knowledge, line-wise scanning-based super-resolution (LSSR) imaging method. To reduce point spread functions overlapping among pixels, we specifically present a super-resolution (SR) imaging architecture to capture a series of low-resolution images using a line-based optical multiplexing technique, which is able to achieve a good balance between imaging quality and speed. In addition, we propose an efficient joint reconstruction algorithm based on total variation and low-rank constraints to generate a high-resolution image from these low-resolution images that contain different spatial details. Meanwhile, existing stripe noises are efficiently suppressed. Experiments on real data show that LSSR imaging has significant advantages over other state-of-the-art methods in terms of visual quality and quantitative measurement.

Real-Time and Accurate UAV Pedestrian Detection for Social Distancing Monitoring in COVID-19 Pandemic.

Coronavirus Disease 2019 (COVID-19) is a highly infectious virus that has created a health crisis for people all over the world. Social distancing has proved to be an effective non-pharmaceutical measure to slow down the spread of COVID-19. As unmanned aerial vehicle (UAV) is a flexible mobile platform, it is a promising option to use UAV for social distance monitoring. Therefore, we propose a lightweight pedestrian detection network to accurately detect pedestrians by human head detection in real-time and then calculate the social distancing between pedestrians on UAV images. In particular, our network follows the PeleeNet as backbone and further incorporates the multi-scale features and spatial attention to enhance the features of small objects, like human heads. The experimental results on Merge-Head dataset show that our method achieves 92.22% AP (average precision) and 76 FPS (frames per second), outperforming YOLOv3 models and SSD models and enabling real-time detection in actual applications. The ablation experiments also indicate that multi-scale feature and spatial attention significantly contribute the performance of pedestrian detection. The test results on UAV-Head dataset show that our method can also achieve high precision pedestrian detection on UAV images with 88.5% AP and 75 FPS. In addition, we have conducted a precision calibration test to obtain the transformation matrix from images (vertical images and tilted images) to real-world coordinate. Based on the accurate pedestrian detection and the transformation matrix, the social distancing monitoring between individuals is reliably achieved.

Oxidative Cleavage-Based Three-Dimensional DNA Biosensor for Ratiometric Detection of Hypochlorous Acid and Myeloperoxidase.

Methods to detect and quantify disease biomarkers with high specificity and sensitivity in biological fluids play a key role in enabling clinical diagnosis, including point-of-care testing. Myeloperoxidase (MPO) is an emerging biomarker for the detection of inflammation, neurodegenerative diseases, and cardiovascular disease, where excess MPO can lead to oxidative damage to biomolecules in homeostatic systems. While numerous methods have been developed for MPO analysis, most techniques are challenging in clinical applications due to the lack of amplification methods, high cost, or other practical drawbacks. Enzyme-linked immunosorbent assays are currently used for the quantification of MPO in clinical practice, which is often limited by the availability of antibodies with high affinity and specificity and the significant nonspecific binding of antibodies to the analytical surface. In contrast, nucleic acid-based biosensors are of interest because of their simplicity, fast response time, low cost, high sensitivity, and low background signal, but detection targets are limited to nucleic acids and non-nucleic acid biomarkers are rare. Recent studies reveal that the modification of a genome in the form of phosphorothioate is specifically sensitive to the oxidative effects of the MPO/H2O2/Cl- system. We developed an oxidative cleavage-based three-dimensional DNA biosensor for rapid, ratiometric detection of HOCl and MPO in a "one-pot" method, which is simple, stable, sensitive, specific, and time-saving and does not require a complex reaction process, such as PCR and enzyme involvement. The constructed biosensor has also been successfully used for MPO detection in complex samples. This strategy is therefore of great value in disease diagnosis and biomedical research.

Polarization prior to single-photon counting image denoising.

Single-photon counting (SPC) imaging technique, which can detect targets in extremely low light levels, has attracted considerable research interest in recent years. To reduce the influence of noise under the low light condition, traditional approaches typically seek various priors from images themselves to construct denoising models, leading to inferior performance as the signal and noise cannot be efficiently distinguished. To address this challenging problem, in this study we propose a novel polarization prior to SPC image denoising based on the observation that a special polarization SPC (PSPC) image has a higher SNR than the SPC image. It enables us to construct a polarization prior to the PSPC image that can transfer efficient targets' spatial details to the denoised SPC image, and hence improves the denoising performance. Specifically, we group similar patches of the PSPC image to form 'anti-noise' dictionaries with high SNR. Then we construct a non-local prior-oriented sparse representation constraint based on the fact that each noisy patch of the SPC image can be sparsely represented by the corresponding 'anti-noise' dictionary. According to this sparse representation constraint, we further formulate an SPC image denoising model by incorporating two terms, i.e., a negative Poisson log-likelihood function for preserving the data fidelity and a total variation constraint to reduce the influence of noise, which is solved by an efficient variable splitting method. In the experiment, we have verified the effectiveness of the proposed method from simulated and real data in terms of visual comparison and quantitative analysis, respectively.

Multi-depth photon-counting imaging based on polarisation modulation.

We present a method for multi-depth imaging that uses polarisation modulation to reconstruct multiple depths from photon-counting observations. The echo photon signals of the scene with multiple targets are modelled under the low-flux condition based on a confocal scanning system. We establish a computational method by constructing the relationship between the received photon rate after polarisation modulation and several variables described for multi-echo signals: the detected average number of photons before polarisation modulation and the flight time corresponding to phase shift based on polarisation modulation. The photon rate is directly calculated from the photon-counting value received by a photon-counting detector based on the Poisson negative log-likelihood function. We suggest solving the abovementioned relationship by changing the waveform of the operating voltage applied to the modulator to derive analytic forms of the detected average number of photons and the flight time. We experimentally demonstrate that the proposed method can accurately reconstruct the reflectivity and depth images of two targets in a 3D scene with a significant improvement beyond what the polarisation modulation imaging method for single-depth can achieve. Moreover, we demonstrate the effectiveness of the proposed method by varying the reflection properties of the targets, the number of echo photons (e.g. less than one photon in a pulse) and the background noise. We exhibit a fast imaging capability with the imaging frequency of one pixel at 8 kHz and the root mean-square error of depth smaller than 6 cm.

DNA nanolantern-based split aptamer probes for in situ ATP imaging in living cells and lighting up mitochondria.

Accurate and specific analysis of adenosine triphosphate (ATP) expression levels in living cells can provide valuable information for understanding cell metabolism, physiological activities and pathologic mechanisms. Herein, DNA nanolantern-based split aptamer nanoprobes are prepared and demonstrated to work well for in situ analysis of ATP expression in living cells. The nanoprobes, which carry multiple split aptamer units on the surface, are easily and inexpensively prepared by a "one-pot" assembly reaction of four short oligonucleotide strands. A series of characterization experiments verify that the nanoprobes have good monodispersity, strong biostability, high cell internalization efficiency, and fluorescence resonance energy transfer (FRET)-based ratiometric response to ATP in the concentration range covering the entire intracellular ATP expression level. By changing the intracellular ATP level via different treatments, the nanoprobes are demonstrated to show excellent performance in intracellular ATP expression analysis, giving a highly ATP concentration-dependent ratiometric fluorescence signal output. ATP-induced formation of large-sized DNA aggregates not only amplifies the FRET signal output, but also makes in situ ATP-imaging analysis in living cells possible. In situ responsive crosslinking of nanoprobes also makes them capable of lighting up the mitochondria of living cells. By simply changing the split aptamer sequence, the proposed DNA nanolantern-based split aptamer strategy might be easily extended to other targets.

Polybrominated Diphenyl Ethers and Heavy Metals in a Regulated E-Waste Recycling Site, Eastern China: Implications for Risk Management.

Serious pollution of multiple chemicals in irregulated e-waste recycling sites (IR-sites) were extensively investigated. However, little is known about the pollution in regulated sites. This study investigated the occurrence of 21 polybrominated diphenyl ethers (PBDEs) and 10 metals in a regulated site, in Eastern China. The concentrations of PBDEs and Cd, Cu, Pb, Sb, and Zn in soils and sediments were 1-4 and 1-3 orders of magnitude lower than those reported in the IR-sites, respectively. However, these were generally comparable to those in the urban and industrial areas. In general, a moderate pollution of PBDEs and metals was present in the vegetables in this area. A health risk assessment model was used to calculate human exposure to metals in soils. The summed non-carcinogenic risks of metals and PBDEs in the investigated soils were 1.59-3.27 and 0.25-0.51 for children and adults, respectively. Arsenic contributed to 47% of the total risks and As risks in 71.4% of the total soil samples exceeded the acceptable level. These results suggested that the pollution from e-waste recycling could be substantially decreased by the regulated activities, relative to poorly controlled operations, but arsenic pollution from the regulated cycling should be further controlled.

A Foreground-Aware Framework for Local Face Attribute Transfer.

In the context of social media, large amounts of headshot photos are taken everyday. Unfortunately, in addition to laborious editing and modification, creating a visually compelling photographic masterpiece for sharing requires advanced professional skills, which are difficult for ordinary Internet users. Though there are many algorithms automatically and globally transferring the style from one image to another, they fail to respect the semantics of the scene and are unable to allow users to merely transfer the attributes of one or two face organs in the foreground region leaving the background region unchanged. To overcome this problem, we developed a novel framework for semantically meaningful local face attribute transfer, which can flexibly transfer the local attribute of a face organ from the reference image to a semantically equivalent organ in the input image, while preserving the background. Our method involves warping the reference photo to match the shape, pose, location, and expression of the input image. The fusion of the warped reference image and input image is then taken as the initialized image for a neural style transfer algorithm. Our method achieves better performance in terms of inception score (3.81) and Fréchet inception distance (80.31), which is about 10% higher than those of competitors, indicating that our framework is capable of producing high-quality and photorealistic attribute transfer results. Both theoretical findings and experimental results are provided to demonstrate the efficacy of the proposed framework, reveal its superiority over other state-of-the-art alternatives.

Unsupervised Exemplar-Domain Aware Image-to-Image Translation.

Image-to-image translation is used to convert an image of a certain style to another of the target style with the original content preserved. A desired translator should be capable of generating diverse results in a controllable many-to-many fashion. To this end, we design a novel deep translator, namely exemplar-domain aware image-to-image translator (EDIT for short). From a logical perspective, the translator needs to perform two main functions, i.e., feature extraction and style transfer. With consideration of logical network partition, the generator of our EDIT comprises of a part of blocks configured by shared parameters, and the rest by varied parameters exported by an exemplar-domain aware parameter network, for explicitly imitating the functionalities of extraction and mapping. The principle behind this is that, for images from multiple domains, the content features can be obtained by an extractor, while (re-)stylization is achieved by mapping the extracted features specifically to different purposes (domains and exemplars). In addition, a discriminator is equipped during the training phase to guarantee the output satisfying the distribution of the target domain. Our EDIT can flexibly and effectively work on multiple domains and arbitrary exemplars in a unified neat model. We conduct experiments to show the efficacy of our design, and reveal its advances over other state-of-the-art methods both quantitatively and qualitatively.

A Generative Adversarial Network for Infrared and Visible Image Fusion Based on Semantic Segmentation.

This paper proposes a new generative adversarial network for infrared and visible image fusion based on semantic segmentation (SSGAN), which can consider not only the low-level features of infrared and visible images, but also the high-level semantic information. Source images can be divided into foregrounds and backgrounds by semantic masks. The generator with a dual-encoder-single-decoder framework is used to extract the feature of foregrounds and backgrounds by different encoder paths. Moreover, the discriminator's input image is designed based on semantic segmentation, which is obtained by combining the foregrounds of the infrared images with the backgrounds of the visible images. Consequently, the prominence of thermal targets in the infrared images and texture details in the visible images can be preserved in the fused images simultaneously. Qualitative and quantitative experiments on publicly available datasets demonstrate that the proposed approach can significantly outperform the state-of-the-art methods.