Video camouflaged object detection(VCOD)has become a fundamental task in computer vision that has attracted significant attention in recent years.Unlike image camouflaged object detection(ICOD),VCOD not only requires ...Video camouflaged object detection(VCOD)has become a fundamental task in computer vision that has attracted significant attention in recent years.Unlike image camouflaged object detection(ICOD),VCOD not only requires spatial cues but also needs motion cues.Thus,effectively utilizing spatiotemporal information is crucial for generating accurate segmentation results.Current VCOD methods,which typically focus on exploring motion representation,often ineffectively integrate spatial and motion features,leading to poor performance in diverse scenarios.To address these issues,we design a novel spatiotemporal network with an encoder-decoder structure.During the encoding stage,an adjacent space-time memory module(ASTM)is employed to extract high-level temporal features(i.e.,motion cues)from the current frame and its adjacent frames.In the decoding stage,a selective space-time aggregation module is introduced to efficiently integrate spatial and temporal features.Additionally,a multi-feature fusion module is developed to progressively refine the rough prediction by utilizing the information provided by multiple types of features.Furthermore,we incorporate multi-task learning into the proposed network to obtain more accurate predictions.Experimental results show that the proposed method outperforms existing cutting-edge baselines on VCOD benchmarks.展开更多
Camouflaged people are extremely expert in actively concealing themselves by effectively utilizing cover and the surrounding environment. Despite advancements in optical detection capabilities through imaging systems,...Camouflaged people are extremely expert in actively concealing themselves by effectively utilizing cover and the surrounding environment. Despite advancements in optical detection capabilities through imaging systems, including spectral, polarization, and infrared technologies, there is still a lack of effective real-time method for accurately detecting small-size and high-efficient camouflaged people in complex real-world scenes. Here, this study proposes a snapshot multispectral image-based camouflaged detection model, multispectral YOLO(MS-YOLO), which utilizes the SPD-Conv and Sim AM modules to effectively represent targets and suppress background interference by exploiting the spatial-spectral target information. Besides, the study constructs the first real-shot multispectral camouflaged people dataset(MSCPD), which encompasses diverse scenes, target scales, and attitudes. To minimize information redundancy, MS-YOLO selects an optimal subset of 12 bands with strong feature representation and minimal inter-band correlation as input. Through experiments on the MSCPD, MS-YOLO achieves a mean Average Precision of 94.31% and real-time detection at 65 frames per second, which confirms the effectiveness and efficiency of our method in detecting camouflaged people in various typical desert and forest scenes. Our approach offers valuable support to improve the perception capabilities of unmanned aerial vehicles in detecting enemy forces and rescuing personnel in battlefield.展开更多
The efficacy of focus-on-form(FonF)within the context of communicatively-oriented language activities is measured via uptake.Uptake is defined as learners’verbal responses immediately following either preemptive or r...The efficacy of focus-on-form(FonF)within the context of communicatively-oriented language activities is measured via uptake.Uptake is defined as learners’verbal responses immediately following either preemptive or reactive FonF instruction(Loewen,2004).The present study investigated what is(not)meant and(not)measured through this definition of uptake.Drawing on the audio-recorded analysis of 20 hours of communicatively–oriented interactions in an intermediate IELTS class with two teachers,this study investigates the frequency of preemptive and reactive incidental FonF,and the subsequent occurrence of uptake in an English as a foreign language context.This study also provided an in-depth qualitative analysis of these classes through field notes,learner notes,and video-recorded data to explore the instances of uptake moves that were not captured through audio-recorded data.The quantitative findings of this study demonstrated a very low and disappointing uptake rate.Furthermore,the study did not find a significant difference between reactive and preemptive FonF in terms of uptake rate.Nonetheless,the qualitative data revealed a myriad of uptake instances not observable via the initial data analysis.Based on these findings,a new definition of uptake is suggested,which includes camouflaged uptake and learners’immediate oral responses to FonF.Since uptake is used to gauge the efficacy of incidental FonF in primarily meaning–oriented classes,it is concluded that audio-recorded data just show the tip of the iceberg as far as the uptake rate is concerned.Thus,second language acquisition researchers are recommended to employ multiple indices to examine the effectiveness of FonF instruction.展开更多
The object detectors can precisely detect the camouflaged object beyond human perception.The investigations reveal that the CNNs-based(Convolution Neural Networks)detectors are vulnerable to adversarial attacks.Some w...The object detectors can precisely detect the camouflaged object beyond human perception.The investigations reveal that the CNNs-based(Convolution Neural Networks)detectors are vulnerable to adversarial attacks.Some works can fool detectors by crafting the adversarial camouflage attached to the object,leading to wrong prediction.It is hard for military operations to utilize the existing adversarial camouflage due to its conspicuous appearance.Motivated by this,this paper proposes the Dual Attribute Adversarial Camouflage(DAAC)for evading the detection by both detectors and humans.Generating DAAC includes two steps:(1)Extracting features from a specific type of scene to generate individual soldier digital camouflage;(2)Attaching the adversarial patch with scene features constraint to the individual soldier digital camouflage to generate the adversarial attribute of DAAC.The visual effects of the individual soldier digital camouflage and the adversarial patch will be improved after integrating with the scene features.Experiment results show that objects camouflaged by DAAC are well integrated with background and achieve visual concealment while remaining effective in fooling object detectors,thus evading the detections by both detectors and humans in the digital domain.This work can serve as the reference for crafting the adversarial camouflage in the physical world.展开更多
Semiconducting conjugated polymer nanoparticles(SPNs)represent an emerging class of phototheranostic materi-als with great promise for cancer treatment.In this report,low-bandgap electron donoracceptor(DA)-conjugated ...Semiconducting conjugated polymer nanoparticles(SPNs)represent an emerging class of phototheranostic materi-als with great promise for cancer treatment.In this report,low-bandgap electron donoracceptor(DA)-conjugated SPNs with sur-face cloaked by red blood cell membrane(RBCM)are developed for highly e ective photoacoustic imaging and photothermal therapy.The resulting RBCM-coated SPN(SPN@RBCM)displays remarkable near-infrared light absorption and good photosta-bility,as well as high photothermal conver-sion e ciency for photoacoustic imaging and photothermal therapy.Particularly,due to the small size(<5 nm),SPN@RBCM has the advantages of deep tumor penetration and rapid clearance from the body with no appreciable toxicity.The RBCM endows the SPNs with prolonged systematic circulation time,less reticuloendothelial system uptake and reduced immune-recognition,hence improving tumor accumulation after intravenous injection,which provides strong photoacoustic signals and exerts excellent photothermal therapeutic e ects.Thus,this work provides a valuable paradigm for safe and highly e cient tumor pho-toacoustic imaging and photothermal therapy for further clinical translation.展开更多
To adapt to a complex and variable environment,self-adaptive camouflage technology is becoming more and more important in all kinds of military applications by overcoming the weakness of the static camouflage.In natur...To adapt to a complex and variable environment,self-adaptive camouflage technology is becoming more and more important in all kinds of military applications by overcoming the weakness of the static camouflage.In nature,the chameleon can achieve self-adaptive camouflage by changing its skin color in real time with the change of the background color.To imitate the chameleon skin,a camouflaged film controlled by a color-changing microfluidic system is proposed in this paper.The film with microfluidic channels fabricated by soft materials can achieve dynamic cloaking and camouflage by circulating color liquids through channels inside the film.By sensing and collecting environmental color change information,the control signal of the microfluidic system can be adjusted in real time to imitate chameleon skin.The microstructure of the film and the working principle of the microfluidic color-changing system are introduced.The mechanism to generate the control signal by information processing of background colors is illustrated.“Canny”double-threshold edge detection algorithm and color similarity are used to analyze and evaluate the camouflage.The tested results show that camouflaged images have a relatively high compatibility with environmental backgrounds and the dynamic cloaking eff ect can be achieved.展开更多
Enhancing the active tumor targeting ability and decreasing the clearance of reticuloendothelial system(RES)are important issues for drug delivery systems(DDSs)in cancer therapy.In recent years,cell membrane camouflag...Enhancing the active tumor targeting ability and decreasing the clearance of reticuloendothelial system(RES)are important issues for drug delivery systems(DDSs)in cancer therapy.In recent years,cell membrane camouflage,as one of the biomimetic modification strategies,has shown huge potential.Many natural properties of source cells can be inherited,allowing the DDSs to successfully avoid phagocytosis by macrophages,prolong circulation time,and achieve homologous targeting to lesion tissue.In this study,a cancer cell membrane camouflaged nanoplatform based on gelatin with a typical core-shell structure was developed for cancer chemotherapy.Doxorubicin(DOX)loaded gelatin nanogel(NG@DOX)acted as the inner core,and 4T1(mouse breast carcinoma cell)membrane was set as the outer shell(M-NG@DOX).The M-NG platform enhanced the ability of homologous targeting due to the surface protein of cell membrane being completely retained,which could promote the cell uptake of homotypic cells,avoid phagocytosis by RAW264.7 macrophages,and therefore increase accumulation in tumor tissue.Meanwhile,due to the better controlled drug release capability of M-NG@DOX,premature release of DOX in circulation could be reduced,minimizing side effects in common chemotherapy.As a result,the biomimetic nanoplatform in this study,obtained by a cancer cell membrane camouflaged drug delivery system,efficiently reached desirable tumor elimination,providing a significant strategy for effective targeted therapy and specific carcinoma therapy.展开更多
Camouflaged object detection(COD)refers to the task of identifying and segmenting objects that blend seamlessly into their surroundings,posing a significant challenge for computer vision systems.In recent years,COD ha...Camouflaged object detection(COD)refers to the task of identifying and segmenting objects that blend seamlessly into their surroundings,posing a significant challenge for computer vision systems.In recent years,COD has garnered widespread attention due to its potential applications in surveillance,wildlife conservation,autonomous systems,and more.While several surveys on COD exist,they often have limitations in terms of the number and scope of papers covered,particularly regarding the rapid advancements made in the field since mid-2023.To address this void,we present the most comprehensive review of COD to date,encompassing both theoretical frameworks and practical contributions to the field.This paper explores various COD methods across four domains,including both image-level and video-level solutions,from the perspectives of traditional and deep learning approaches.We thoroughly investigate the correlations between COD and other camouflaged scenario methods,thereby laying the theoretical foundation for subsequent analyses.Furthermore,we delve into novel tasks such as referring-based COD and collaborative COD,which have not been fully addressed in previous works.Beyond object-level detection,we also summarize extended methods for instance-level tasks,including camouflaged instance segmentation,counting,and ranking.Additionally,we provide an overview of commonly used benchmarks and evaluation metrics in COD tasks,conducting a comprehensive evaluation of deep learning-based techniques in both image and video domains,considering both qualitative and quantitative performance.Finally,we discuss the limitations of current COD models and propose 9 promising directions for future research,focusing on addressing inherent challenges and exploring novel,meaningful technologies.This comprehensive examination aims to deepen the understanding of COD models and related methods in camouflaged scenarios.For those interested,a curated list of CODrelated techniques,datasets,and additional resources can be found at https://github.com/ChunmingHe/awesome-concealed-objectsegmentation.展开更多
This paper introduces deep gradient network(DGNet),a novel deep framework that exploits object gradient supervision for camouflaged object detection(COD).It decouples the task into two connected branches,i.e.,a contex...This paper introduces deep gradient network(DGNet),a novel deep framework that exploits object gradient supervision for camouflaged object detection(COD).It decouples the task into two connected branches,i.e.,a context and a texture encoder.The es-sential connection is the gradient-induced transition,representing a soft grouping between context and texture features.Benefiting from the simple but efficient framework,DGNet outperforms existing state-of-the-art COD models by a large margin.Notably,our efficient version,DGNet-S,runs in real-time(80 fps)and achieves comparable results to the cutting-edge model JCSOD-CVPR21 with only 6.82%parameters.The application results also show that the proposed DGNet performs well in the polyp segmentation,defect detec-tion,and transparent object segmentation tasks.The code will be made available at https://github.com/GewelsJI/DGNet.展开更多
Cell membrane camouflaged nanoparticles have been widely used in the field of drug leads discovery attribute to their unique biointerface targeting function.However,random orientation of cell membrane coating does not...Cell membrane camouflaged nanoparticles have been widely used in the field of drug leads discovery attribute to their unique biointerface targeting function.However,random orientation of cell membrane coating does not guarantee effective and appropriate binding of drugs to specific sites,especially when applied to intracellular regions of transmembrane proteins.Bioorthogonal reactions have been rapidly developed as a specific and reliable method for cell membrane functionalization without disturbing living biosystem.Herein,inside-out cell membrane camouflaged magnetic nanoparticles(IOCMMNPs)were accurately constructed via bioorthogonal reactions to screen small molecule inhibitors targeting intracellular tyrosine kinase domain of vascular endothelial growth factor recptor-2.Azide functionalized cell membrane acted as a platform for specific covalently coupling with alkynyl functionalized magnetic Fe_(3)O_(4)nanoparticles to prepare IOCMMNPs.The inside-out orientation of cell membrane was successfully verified by immunogold staining and sialic acid quantification assay.Ultimately,two compounds,senkyunolide A and ligustilidel,were successfully captured,and their potential antiproliferative activities were further testified by pharmacological experiments.It is anticipated that the proposed inside-out cell membrane coating strategy endows tremendous versatility for engineering cell membrane camouflaged nanoparticles and promotes the development of drug leads discovery platforms.展开更多
Camouflaged targets are a type of nonsalient target with high foreground and background fusion and minimal target feature information,making target recognition extremely difficult.Most detection algorithms for camoufl...Camouflaged targets are a type of nonsalient target with high foreground and background fusion and minimal target feature information,making target recognition extremely difficult.Most detection algorithms for camouflaged targets use only the target’s single-band information,resulting in low detection accuracy and a high missed detection rate.We present a multimodal image fusion camouflaged target detection technique (MIF-YOLOv5) in this paper.First,we provide a multimodal image input to achieve pixel-level fusion of the camouflaged target’s optical and infrared images to improve the effective feature information of the camouflaged target.Second,a loss function is created,and the K-Means++clustering technique is used to optimize the target anchor frame in the dataset to increase camouflage personnel detection accuracy and robustness.Finally,a comprehensive detection index of camouflaged targets is proposed to compare the overall effectiveness of various approaches.More crucially,we create a multispectral camouflage target dataset to test the suggested technique.Experimental results show that the proposed method has the best comprehensive detection performance,with a detection accuracy of 96.5%,a recognition probability of92.5%,a parameter number increase of 1×10^(4),a theoretical calculation amount increase of 0.03 GFLOPs,and a comprehensive detection index of 0.85.The advantage of this method in terms of detection accuracy is also apparent in performance comparisons with other target algorithms.展开更多
Although nano-immunotherapy has advanced dramatically in recent times,there remain two significant hurdles related to immune systems in cancer treatment,such as(namely)inevitable immune elimination of nanoplat-forms a...Although nano-immunotherapy has advanced dramatically in recent times,there remain two significant hurdles related to immune systems in cancer treatment,such as(namely)inevitable immune elimination of nanoplat-forms and severely immunosuppressive microenvironment with low immunogenicity,hampering the perfor-mance of nanomedicines.To address these issues,several immune-regulating camouflaged nanocomposites have emerged as prevailing strategies due to their unique characteristics and specific functionalities.In this review,we emphasize the composition,performances,and mechanisms of various immune-regulating camouflaged nano-platforms,including polymer-coated,cell membrane-camouflaged,and exosome-based nanoplatforms to evade the immune clearance of nanoplatforms or upregulate the immune function against the tumor.Further,we discuss the applications of these immune-regulating camouflaged nanoplatforms in directly boosting cancer immunotherapy and some immunogenic cell death-inducing immunotherapeutic modalities,such as chemo-therapy,photothermal therapy,and reactive oxygen species-mediated immunotherapies,highlighting the cur-rent progress and recent advancements.Finally,we conclude the article with interesting perspectives,suggesting future tendencies of these innovative camouflaged constructs towards their translation pipeline.展开更多
Surfaces with micro-nanoscale structures show different optical responses,including infrared reflection,thermal radiation,and protective coloration.Direct realization of structure camouflage is important for material ...Surfaces with micro-nanoscale structures show different optical responses,including infrared reflection,thermal radiation,and protective coloration.Direct realization of structure camouflage is important for material functionalities.However,external cloaks or coatings are necessary in structure camouflage,which limits the surface functionality.Here,we propose a novel strategy for the direct structure camouflage through topography inherited removal(TIR)with ultrafast laser,featuring pristine topography preservation and scattering surface fabrication.After multistep TIR,pristine topographies are partially and uniformly removed to preserve the original designed structures.Optical response changes show the suppression of specular reflection by uniformizing reflected light intensity to a low level on the inherited surface.We produce various structure camouflages on large scaled substrates,and demonstrate applications of information encryption in code extraction and word recognition through structure camouflage.The proposed strategy opens opportunities for infrared camouflage and other technologies,such as thermal management,device security,and information encryption.展开更多
Integrated circuit （IC） camouflaging technique has been applied as a countermeasure against reverse engineering （RE）. However, its effectiveness is threatened by a boolean satisfiability （SAT） based de-camouflag...Integrated circuit （IC） camouflaging technique has been applied as a countermeasure against reverse engineering （RE）. However, its effectiveness is threatened by a boolean satisfiability （SAT） based de-camouflaging attack, which is able to restore the camouflaged circuit within only minutes. As a defense to the SAT-based de-camouflaging attack, a brand new camouflaging strategy （called CamoPerturb） has been proposed recently, which perturbs one minterm by changing one gate＇s functionality and then restores the perturbed circuit with a separated camouflaged block, achieving good resistance against the SAT-based attack. In this paper, we analyze the security vulnerabilities of CamoPerturb by illustrating the mechanism of minterm perturbation induced by gate replacement, then propose an attack to restore the changed gate＇s functionality, and recover the camouflaged circuit. The attack algorithm is facilitated by sensitization and implication principles in automatic test pattern generation （ATPG） techniques. Experimental results demonstrate that our method is able to restore the camouflaged circuits with very little time consumption.展开更多
The combination of advanced photoelectric detectors has rendered single-band camouflage materials ineffective,necessitating the development of infrared multispectral camouflage.However,the design and fabrication of ex...The combination of advanced photoelectric detectors has rendered single-band camouflage materials ineffective,necessitating the development of infrared multispectral camouflage.However,the design and fabrication of existing works remain complex as they usually require the integration of multiscale structures.Here,we introduce phase modulation into the infrared camouflage metasurfaces with metal-dielectric-metal configuration,enabling them to achieve camouflage across more bands.Based on this strategy,a simple but effective single-layer cascaded metasurface is demonstrated for the first time to achieve low reflection at multi-wavelength lasers,low infrared radiation in atmospheric windows,and broadband thermal management.As a proof-of-concept,a 4-inch sample with a minimum linewidth of 1.8μm is fabricated using photolithography.The excellent infrared multispectral camouflage performance is verified in experiments,showing low reflectance in 0.9–1.6μm,low infrared emissivity in mid-wavelength infrared(MWIR)and long-wavelength infrared(LWIR)bands,and high absorptance at the wavelength of 10.6μm.Meanwhile,broadband high emissivity in 5–8μm can provide high-performance radiative heat dissipation.When the input power is 1.57 W·cm^(-2),the surface/radiation temperature of the metasurface decreases by 5.3℃/18.7℃ compared to the reference.The proposed metasurface may trigger further innovation in the design and application of compact multispectral optical devices.展开更多
A laser-induced periodic surface structure(LIPSS),which can be easily produced by femtosecond laser ablation,is a unique nanostructure with a visible refractive color that can be controlled by altering its orientation...A laser-induced periodic surface structure(LIPSS),which can be easily produced by femtosecond laser ablation,is a unique nanostructure with a visible refractive color that can be controlled by altering its orientation and uniformity,making it suitable for use in colorful marking,camouflage,and anticounterfeiting measures.However,single-mode information camouflage can no longer meet increasingly higher-level security requirements.Therefore,metasurfaces offer revolutionary solutions.In this study,conceptual metasurfaces of pure tungsten are theoretically proposed and verified using hierarchical LIPSS/nanoparticle(NP)nanostructures as meta-atoms.The anisotropy of the LIPSS nanostructure enables polarization-sensitive optical modulation,whereas the spatial configuration,NPs size,and period of LIPSS in the LIPSS/NP meta-atoms provide flexibility for tailoring broadband optical responses.In xpolarization,the LIPSS/NP meta-atom system provides more visible colors and divergent infrared absorption(emission)than in y-polarized and unpolarized modes,paving the way for vividly colorful polarization-sensitive displays and information camouflage in infrared bands.A simplified rendition of the world-famous painting“The Starry Night”by Van Gogh is used as a proof-of-concept.Preliminary experimental results are presented,based on which the feasibility and challenges for laser nanomanufacturing of the proposed conceptual metasurfaces are discussed,aiming to provide inspiration for the development of novel metasurfaces through interdisciplinary studies.展开更多
Li Jiayue,a talented artist from Beichuan Qiang Autonomous County,China,has captivated(迷住)audiences with his remarkable optical illusions.Although he initially studied electrical automation,Li's passion for art ...Li Jiayue,a talented artist from Beichuan Qiang Autonomous County,China,has captivated(迷住)audiences with his remarkable optical illusions.Although he initially studied electrical automation,Li's passion for art led him to a career dedicated to creating stunning visual illusions.Using his exceptional painting skills,Li camouflages objects like lamp posts,tree trunks and even large buildings,blending them seamlessly with their surroundings in a way that confuses the eye and captivates the mind.He meticulously studies the textures,colors and patterns of the environment,ensuring that each stroke of his brush perfectly matches the background.This attention to detail allows him to create illusions that are not only visually striking but also incredibly convincing,making it difficult for viewers to distinguish between what is real and what is an illusion.展开更多
Combining deep-learning image inpainting algorithms with the microfluidic technology,the paper proposes a method to achieve dynamic stealth and camouflage by using a microfluidic vision camouflage system simulating th...Combining deep-learning image inpainting algorithms with the microfluidic technology,the paper proposes a method to achieve dynamic stealth and camouflage by using a microfluidic vision camouflage system simulating the chameleon skin.The basic principle is to perceive color changes in the external environment and collect ambient image information,and then utilize the image inpainting algorithm to adjust the control signals of the microfluidic system in real time.The detailed working principle of the microfluidic vision camouflage system is presented,and the mechanism of generating control signals for the system through deep-learning image inpainting algorithms and image-processing techniques is elucidated.The camouflage effect of the chameleon skin is analyzed and evaluated using color similarity.Results indicate that the camouflaged images are consistent with the background environment,thereby improving the target’s stealth and maneuvering characteristics.The camouflage technology developed in the paper based on the microfluidic vision camouflage system can be applied to several situations,such as military camouflage uniforms,robot skins,and weapon equipment.展开更多
This study serves as a guide to the development of a polydimethylsiloxane(PDMS)-encapsulated liquid metal-MXene aerogel,which has proven to be highly effective for electromagnetic wave absorption,particularly in salin...This study serves as a guide to the development of a polydimethylsiloxane(PDMS)-encapsulated liquid metal-MXene aerogel,which has proven to be highly effective for electromagnetic wave absorption,particularly in saline environments.Through directional freezing and casting techniques,we have optimized the sample to exhibit enhanced absorption properties,achieving a reflection loss peak of-63.10 dB at 14.36 GHz.Variations in liquid metal content were found to significantly impact the complex permittivity of the aerogel,resulting in decreases observed in both real and imaginary components.This underscores the crucial role of conductivity in electromagnetic wave damping.Simultaneously,increases in tangent loss and attenuation constant highlight the vital contribution of MXene towards dissipating electromagnetic energy.Our best sample exhibits enhanced mechanical robustness,as evidenced by a high tensile modulus of 1 MPa.Notably,this exceptional performance is sustained for an extended period of 4 weeks even under harsh conditions such as high temperature,acid mist exposure,alkaline exposure,and immersion in synthetic seawater.By testing the thermal camouflage performance,samples achieved processable and efficient camouflage performance at multiple temperatures.This comprehensive dataset confirms the adaptability of the PDMS-encapsulated liquid metal-MXene aerogel as an effective solution for electromagnetic wave absorption in challenging environmental scenarios.展开更多
Deep neural networks,especially face recognition models,have been shown to be vulnerable to adversarial examples.However,existing attack methods for face recognition systems either cannot attack black-box models,are n...Deep neural networks,especially face recognition models,have been shown to be vulnerable to adversarial examples.However,existing attack methods for face recognition systems either cannot attack black-box models,are not universal,have cumbersome deployment processes,or lack camouflage and are easily detected by the human eye.In this paper,we propose an adversarial pattern generation method for face recognition and achieve universal black-box attacks by pasting the pattern on the frame of goggles.To achieve visual camouflage,we use a generative adversarial network(GAN).The scale of the generative network of GAN is increased to balance the performance conflict between concealment and adversarial behavior,the perceptual loss function based on VGG19 is used to constrain the color style and enhance GAN’s learning ability,and the fine-grained meta-learning adversarial attack strategy is used to carry out black-box attacks.Sufficient visualization results demonstrate that compared with existing methods,the proposed method can generate samples with camouflage and adversarial characteristics.Meanwhile,extensive quantitative experiments show that the generated samples have a high attack success rate against black-box models.展开更多
文摘Video camouflaged object detection(VCOD)has become a fundamental task in computer vision that has attracted significant attention in recent years.Unlike image camouflaged object detection(ICOD),VCOD not only requires spatial cues but also needs motion cues.Thus,effectively utilizing spatiotemporal information is crucial for generating accurate segmentation results.Current VCOD methods,which typically focus on exploring motion representation,often ineffectively integrate spatial and motion features,leading to poor performance in diverse scenarios.To address these issues,we design a novel spatiotemporal network with an encoder-decoder structure.During the encoding stage,an adjacent space-time memory module(ASTM)is employed to extract high-level temporal features(i.e.,motion cues)from the current frame and its adjacent frames.In the decoding stage,a selective space-time aggregation module is introduced to efficiently integrate spatial and temporal features.Additionally,a multi-feature fusion module is developed to progressively refine the rough prediction by utilizing the information provided by multiple types of features.Furthermore,we incorporate multi-task learning into the proposed network to obtain more accurate predictions.Experimental results show that the proposed method outperforms existing cutting-edge baselines on VCOD benchmarks.
基金support by the National Natural Science Foundation of China (Grant No. 62005049)Natural Science Foundation of Fujian Province (Grant Nos. 2020J01451, 2022J05113)Education and Scientific Research Program for Young and Middleaged Teachers in Fujian Province (Grant No. JAT210035)。
文摘Camouflaged people are extremely expert in actively concealing themselves by effectively utilizing cover and the surrounding environment. Despite advancements in optical detection capabilities through imaging systems, including spectral, polarization, and infrared technologies, there is still a lack of effective real-time method for accurately detecting small-size and high-efficient camouflaged people in complex real-world scenes. Here, this study proposes a snapshot multispectral image-based camouflaged detection model, multispectral YOLO(MS-YOLO), which utilizes the SPD-Conv and Sim AM modules to effectively represent targets and suppress background interference by exploiting the spatial-spectral target information. Besides, the study constructs the first real-shot multispectral camouflaged people dataset(MSCPD), which encompasses diverse scenes, target scales, and attitudes. To minimize information redundancy, MS-YOLO selects an optimal subset of 12 bands with strong feature representation and minimal inter-band correlation as input. Through experiments on the MSCPD, MS-YOLO achieves a mean Average Precision of 94.31% and real-time detection at 65 frames per second, which confirms the effectiveness and efficiency of our method in detecting camouflaged people in various typical desert and forest scenes. Our approach offers valuable support to improve the perception capabilities of unmanned aerial vehicles in detecting enemy forces and rescuing personnel in battlefield.
文摘The efficacy of focus-on-form(FonF)within the context of communicatively-oriented language activities is measured via uptake.Uptake is defined as learners’verbal responses immediately following either preemptive or reactive FonF instruction(Loewen,2004).The present study investigated what is(not)meant and(not)measured through this definition of uptake.Drawing on the audio-recorded analysis of 20 hours of communicatively–oriented interactions in an intermediate IELTS class with two teachers,this study investigates the frequency of preemptive and reactive incidental FonF,and the subsequent occurrence of uptake in an English as a foreign language context.This study also provided an in-depth qualitative analysis of these classes through field notes,learner notes,and video-recorded data to explore the instances of uptake moves that were not captured through audio-recorded data.The quantitative findings of this study demonstrated a very low and disappointing uptake rate.Furthermore,the study did not find a significant difference between reactive and preemptive FonF in terms of uptake rate.Nonetheless,the qualitative data revealed a myriad of uptake instances not observable via the initial data analysis.Based on these findings,a new definition of uptake is suggested,which includes camouflaged uptake and learners’immediate oral responses to FonF.Since uptake is used to gauge the efficacy of incidental FonF in primarily meaning–oriented classes,it is concluded that audio-recorded data just show the tip of the iceberg as far as the uptake rate is concerned.Thus,second language acquisition researchers are recommended to employ multiple indices to examine the effectiveness of FonF instruction.
基金National Natural Science Foundation of China(grant number 61801512,grant number 62071484)Natural Science Foundation of Jiangsu Province(grant number BK20180080)to provide fund for conducting experiments。
文摘The object detectors can precisely detect the camouflaged object beyond human perception.The investigations reveal that the CNNs-based(Convolution Neural Networks)detectors are vulnerable to adversarial attacks.Some works can fool detectors by crafting the adversarial camouflage attached to the object,leading to wrong prediction.It is hard for military operations to utilize the existing adversarial camouflage due to its conspicuous appearance.Motivated by this,this paper proposes the Dual Attribute Adversarial Camouflage(DAAC)for evading the detection by both detectors and humans.Generating DAAC includes two steps:(1)Extracting features from a specific type of scene to generate individual soldier digital camouflage;(2)Attaching the adversarial patch with scene features constraint to the individual soldier digital camouflage to generate the adversarial attribute of DAAC.The visual effects of the individual soldier digital camouflage and the adversarial patch will be improved after integrating with the scene features.Experiment results show that objects camouflaged by DAAC are well integrated with background and achieve visual concealment while remaining effective in fooling object detectors,thus evading the detections by both detectors and humans in the digital domain.This work can serve as the reference for crafting the adversarial camouflage in the physical world.
基金supported by the National Natural Science Foundation of China(Grant Nos.61727823,51873160)the joint research project of Health and Education Commission of Fujian Province(Grant No.2019-WJ-20).
文摘Semiconducting conjugated polymer nanoparticles(SPNs)represent an emerging class of phototheranostic materi-als with great promise for cancer treatment.In this report,low-bandgap electron donoracceptor(DA)-conjugated SPNs with sur-face cloaked by red blood cell membrane(RBCM)are developed for highly e ective photoacoustic imaging and photothermal therapy.The resulting RBCM-coated SPN(SPN@RBCM)displays remarkable near-infrared light absorption and good photosta-bility,as well as high photothermal conver-sion e ciency for photoacoustic imaging and photothermal therapy.Particularly,due to the small size(<5 nm),SPN@RBCM has the advantages of deep tumor penetration and rapid clearance from the body with no appreciable toxicity.The RBCM endows the SPNs with prolonged systematic circulation time,less reticuloendothelial system uptake and reduced immune-recognition,hence improving tumor accumulation after intravenous injection,which provides strong photoacoustic signals and exerts excellent photothermal therapeutic e ects.Thus,this work provides a valuable paradigm for safe and highly e cient tumor pho-toacoustic imaging and photothermal therapy for further clinical translation.
基金the National Natural Science Foundation of China for the support(No.51175101)on this paper.
文摘To adapt to a complex and variable environment,self-adaptive camouflage technology is becoming more and more important in all kinds of military applications by overcoming the weakness of the static camouflage.In nature,the chameleon can achieve self-adaptive camouflage by changing its skin color in real time with the change of the background color.To imitate the chameleon skin,a camouflaged film controlled by a color-changing microfluidic system is proposed in this paper.The film with microfluidic channels fabricated by soft materials can achieve dynamic cloaking and camouflage by circulating color liquids through channels inside the film.By sensing and collecting environmental color change information,the control signal of the microfluidic system can be adjusted in real time to imitate chameleon skin.The microstructure of the film and the working principle of the microfluidic color-changing system are introduced.The mechanism to generate the control signal by information processing of background colors is illustrated.“Canny”double-threshold edge detection algorithm and color similarity are used to analyze and evaluate the camouflage.The tested results show that camouflaged images have a relatively high compatibility with environmental backgrounds and the dynamic cloaking eff ect can be achieved.
基金supported by grants from Science and Technology Program of Guangzhou,China(No.201804010146)Guangzhou Science and Technology Program City-University Joint Funding Project(No.2023A03J0001)。
文摘Enhancing the active tumor targeting ability and decreasing the clearance of reticuloendothelial system(RES)are important issues for drug delivery systems(DDSs)in cancer therapy.In recent years,cell membrane camouflage,as one of the biomimetic modification strategies,has shown huge potential.Many natural properties of source cells can be inherited,allowing the DDSs to successfully avoid phagocytosis by macrophages,prolong circulation time,and achieve homologous targeting to lesion tissue.In this study,a cancer cell membrane camouflaged nanoplatform based on gelatin with a typical core-shell structure was developed for cancer chemotherapy.Doxorubicin(DOX)loaded gelatin nanogel(NG@DOX)acted as the inner core,and 4T1(mouse breast carcinoma cell)membrane was set as the outer shell(M-NG@DOX).The M-NG platform enhanced the ability of homologous targeting due to the surface protein of cell membrane being completely retained,which could promote the cell uptake of homotypic cells,avoid phagocytosis by RAW264.7 macrophages,and therefore increase accumulation in tumor tissue.Meanwhile,due to the better controlled drug release capability of M-NG@DOX,premature release of DOX in circulation could be reduced,minimizing side effects in common chemotherapy.As a result,the biomimetic nanoplatform in this study,obtained by a cancer cell membrane camouflaged drug delivery system,efficiently reached desirable tumor elimination,providing a significant strategy for effective targeted therapy and specific carcinoma therapy.
基金supported by the STI 2030-Major Projects(No.2021ZD0201404).
文摘Camouflaged object detection(COD)refers to the task of identifying and segmenting objects that blend seamlessly into their surroundings,posing a significant challenge for computer vision systems.In recent years,COD has garnered widespread attention due to its potential applications in surveillance,wildlife conservation,autonomous systems,and more.While several surveys on COD exist,they often have limitations in terms of the number and scope of papers covered,particularly regarding the rapid advancements made in the field since mid-2023.To address this void,we present the most comprehensive review of COD to date,encompassing both theoretical frameworks and practical contributions to the field.This paper explores various COD methods across four domains,including both image-level and video-level solutions,from the perspectives of traditional and deep learning approaches.We thoroughly investigate the correlations between COD and other camouflaged scenario methods,thereby laying the theoretical foundation for subsequent analyses.Furthermore,we delve into novel tasks such as referring-based COD and collaborative COD,which have not been fully addressed in previous works.Beyond object-level detection,we also summarize extended methods for instance-level tasks,including camouflaged instance segmentation,counting,and ranking.Additionally,we provide an overview of commonly used benchmarks and evaluation metrics in COD tasks,conducting a comprehensive evaluation of deep learning-based techniques in both image and video domains,considering both qualitative and quantitative performance.Finally,we discuss the limitations of current COD models and propose 9 promising directions for future research,focusing on addressing inherent challenges and exploring novel,meaningful technologies.This comprehensive examination aims to deepen the understanding of COD models and related methods in camouflaged scenarios.For those interested,a curated list of CODrelated techniques,datasets,and additional resources can be found at https://github.com/ChunmingHe/awesome-concealed-objectsegmentation.
文摘This paper introduces deep gradient network(DGNet),a novel deep framework that exploits object gradient supervision for camouflaged object detection(COD).It decouples the task into two connected branches,i.e.,a context and a texture encoder.The es-sential connection is the gradient-induced transition,representing a soft grouping between context and texture features.Benefiting from the simple but efficient framework,DGNet outperforms existing state-of-the-art COD models by a large margin.Notably,our efficient version,DGNet-S,runs in real-time(80 fps)and achieves comparable results to the cutting-edge model JCSOD-CVPR21 with only 6.82%parameters.The application results also show that the proposed DGNet performs well in the polyp segmentation,defect detec-tion,and transparent object segmentation tasks.The code will be made available at https://github.com/GewelsJI/DGNet.
基金the National Natural Science Foundation of China(No.82073807)。
文摘Cell membrane camouflaged nanoparticles have been widely used in the field of drug leads discovery attribute to their unique biointerface targeting function.However,random orientation of cell membrane coating does not guarantee effective and appropriate binding of drugs to specific sites,especially when applied to intracellular regions of transmembrane proteins.Bioorthogonal reactions have been rapidly developed as a specific and reliable method for cell membrane functionalization without disturbing living biosystem.Herein,inside-out cell membrane camouflaged magnetic nanoparticles(IOCMMNPs)were accurately constructed via bioorthogonal reactions to screen small molecule inhibitors targeting intracellular tyrosine kinase domain of vascular endothelial growth factor recptor-2.Azide functionalized cell membrane acted as a platform for specific covalently coupling with alkynyl functionalized magnetic Fe_(3)O_(4)nanoparticles to prepare IOCMMNPs.The inside-out orientation of cell membrane was successfully verified by immunogold staining and sialic acid quantification assay.Ultimately,two compounds,senkyunolide A and ligustilidel,were successfully captured,and their potential antiproliferative activities were further testified by pharmacological experiments.It is anticipated that the proposed inside-out cell membrane coating strategy endows tremendous versatility for engineering cell membrane camouflaged nanoparticles and promotes the development of drug leads discovery platforms.
基金Project supported by the Shandong Provincial Natural Science Foundation of China(No.ZR2020MF015)the Aerospace Science and Technology Innovation Institute Stabilization Support Project(No.ZY0110020009)。
文摘Camouflaged targets are a type of nonsalient target with high foreground and background fusion and minimal target feature information,making target recognition extremely difficult.Most detection algorithms for camouflaged targets use only the target’s single-band information,resulting in low detection accuracy and a high missed detection rate.We present a multimodal image fusion camouflaged target detection technique (MIF-YOLOv5) in this paper.First,we provide a multimodal image input to achieve pixel-level fusion of the camouflaged target’s optical and infrared images to improve the effective feature information of the camouflaged target.Second,a loss function is created,and the K-Means++clustering technique is used to optimize the target anchor frame in the dataset to increase camouflage personnel detection accuracy and robustness.Finally,a comprehensive detection index of camouflaged targets is proposed to compare the overall effectiveness of various approaches.More crucially,we create a multispectral camouflage target dataset to test the suggested technique.Experimental results show that the proposed method has the best comprehensive detection performance,with a detection accuracy of 96.5%,a recognition probability of92.5%,a parameter number increase of 1×10^(4),a theoretical calculation amount increase of 0.03 GFLOPs,and a comprehensive detection index of 0.85.The advantage of this method in terms of detection accuracy is also apparent in performance comparisons with other target algorithms.
基金Financial support from the National Key Research&Development Program of China(2019YFE0113600)National Natural Science Foundation of China(NSFC,81971734,and 32071323)Program for Innovative Research Team in Science and Technology in Fujian Province University,and the Scientific Research Funds of Huaqiao University(20BS104).
文摘Although nano-immunotherapy has advanced dramatically in recent times,there remain two significant hurdles related to immune systems in cancer treatment,such as(namely)inevitable immune elimination of nanoplat-forms and severely immunosuppressive microenvironment with low immunogenicity,hampering the perfor-mance of nanomedicines.To address these issues,several immune-regulating camouflaged nanocomposites have emerged as prevailing strategies due to their unique characteristics and specific functionalities.In this review,we emphasize the composition,performances,and mechanisms of various immune-regulating camouflaged nano-platforms,including polymer-coated,cell membrane-camouflaged,and exosome-based nanoplatforms to evade the immune clearance of nanoplatforms or upregulate the immune function against the tumor.Further,we discuss the applications of these immune-regulating camouflaged nanoplatforms in directly boosting cancer immunotherapy and some immunogenic cell death-inducing immunotherapeutic modalities,such as chemo-therapy,photothermal therapy,and reactive oxygen species-mediated immunotherapies,highlighting the cur-rent progress and recent advancements.Finally,we conclude the article with interesting perspectives,suggesting future tendencies of these innovative camouflaged constructs towards their translation pipeline.
基金supported by the National Natural Science Foundation of China(No.52075289).
文摘Surfaces with micro-nanoscale structures show different optical responses,including infrared reflection,thermal radiation,and protective coloration.Direct realization of structure camouflage is important for material functionalities.However,external cloaks or coatings are necessary in structure camouflage,which limits the surface functionality.Here,we propose a novel strategy for the direct structure camouflage through topography inherited removal(TIR)with ultrafast laser,featuring pristine topography preservation and scattering surface fabrication.After multistep TIR,pristine topographies are partially and uniformly removed to preserve the original designed structures.Optical response changes show the suppression of specular reflection by uniformizing reflected light intensity to a low level on the inherited surface.We produce various structure camouflages on large scaled substrates,and demonstrate applications of information encryption in code extraction and word recognition through structure camouflage.The proposed strategy opens opportunities for infrared camouflage and other technologies,such as thermal management,device security,and information encryption.
基金This work was supported by the National Natural Science Foundation of China under Grant No. 61774091.
文摘Integrated circuit （IC） camouflaging technique has been applied as a countermeasure against reverse engineering （RE）. However, its effectiveness is threatened by a boolean satisfiability （SAT） based de-camouflaging attack, which is able to restore the camouflaged circuit within only minutes. As a defense to the SAT-based de-camouflaging attack, a brand new camouflaging strategy （called CamoPerturb） has been proposed recently, which perturbs one minterm by changing one gate＇s functionality and then restores the perturbed circuit with a separated camouflaged block, achieving good resistance against the SAT-based attack. In this paper, we analyze the security vulnerabilities of CamoPerturb by illustrating the mechanism of minterm perturbation induced by gate replacement, then propose an attack to restore the changed gate＇s functionality, and recover the camouflaged circuit. The attack algorithm is facilitated by sensitization and implication principles in automatic test pattern generation （ATPG） techniques. Experimental results demonstrate that our method is able to restore the camouflaged circuits with very little time consumption.
基金financial supports from the National Natural Science Foundation of China(Grant Nos.51925503&52105575)the Fundamental Research Funds for the Central Universities(Grant No.QTZX23063)+2 种基金the Aeronautical Science Foundation of China(Grant No.2022Z073081001)the Postdoctoral Fellowship Program of CPSF(Grant No.GZC20232028)the Open Research Funds of State Key Laboratory of Intelligent Manufacturing Equipment and Technology(Grant No.IMETKF2024008).
文摘The combination of advanced photoelectric detectors has rendered single-band camouflage materials ineffective,necessitating the development of infrared multispectral camouflage.However,the design and fabrication of existing works remain complex as they usually require the integration of multiscale structures.Here,we introduce phase modulation into the infrared camouflage metasurfaces with metal-dielectric-metal configuration,enabling them to achieve camouflage across more bands.Based on this strategy,a simple but effective single-layer cascaded metasurface is demonstrated for the first time to achieve low reflection at multi-wavelength lasers,low infrared radiation in atmospheric windows,and broadband thermal management.As a proof-of-concept,a 4-inch sample with a minimum linewidth of 1.8μm is fabricated using photolithography.The excellent infrared multispectral camouflage performance is verified in experiments,showing low reflectance in 0.9–1.6μm,low infrared emissivity in mid-wavelength infrared(MWIR)and long-wavelength infrared(LWIR)bands,and high absorptance at the wavelength of 10.6μm.Meanwhile,broadband high emissivity in 5–8μm can provide high-performance radiative heat dissipation.When the input power is 1.57 W·cm^(-2),the surface/radiation temperature of the metasurface decreases by 5.3℃/18.7℃ compared to the reference.The proposed metasurface may trigger further innovation in the design and application of compact multispectral optical devices.
基金financial support received from the Shanghai Pujiang Program(23PJ1406500).
文摘A laser-induced periodic surface structure(LIPSS),which can be easily produced by femtosecond laser ablation,is a unique nanostructure with a visible refractive color that can be controlled by altering its orientation and uniformity,making it suitable for use in colorful marking,camouflage,and anticounterfeiting measures.However,single-mode information camouflage can no longer meet increasingly higher-level security requirements.Therefore,metasurfaces offer revolutionary solutions.In this study,conceptual metasurfaces of pure tungsten are theoretically proposed and verified using hierarchical LIPSS/nanoparticle(NP)nanostructures as meta-atoms.The anisotropy of the LIPSS nanostructure enables polarization-sensitive optical modulation,whereas the spatial configuration,NPs size,and period of LIPSS in the LIPSS/NP meta-atoms provide flexibility for tailoring broadband optical responses.In xpolarization,the LIPSS/NP meta-atom system provides more visible colors and divergent infrared absorption(emission)than in y-polarized and unpolarized modes,paving the way for vividly colorful polarization-sensitive displays and information camouflage in infrared bands.A simplified rendition of the world-famous painting“The Starry Night”by Van Gogh is used as a proof-of-concept.Preliminary experimental results are presented,based on which the feasibility and challenges for laser nanomanufacturing of the proposed conceptual metasurfaces are discussed,aiming to provide inspiration for the development of novel metasurfaces through interdisciplinary studies.
文摘Li Jiayue,a talented artist from Beichuan Qiang Autonomous County,China,has captivated(迷住)audiences with his remarkable optical illusions.Although he initially studied electrical automation,Li's passion for art led him to a career dedicated to creating stunning visual illusions.Using his exceptional painting skills,Li camouflages objects like lamp posts,tree trunks and even large buildings,blending them seamlessly with their surroundings in a way that confuses the eye and captivates the mind.He meticulously studies the textures,colors and patterns of the environment,ensuring that each stroke of his brush perfectly matches the background.This attention to detail allows him to create illusions that are not only visually striking but also incredibly convincing,making it difficult for viewers to distinguish between what is real and what is an illusion.
基金the National Natural Science Foundation of China for the support(No.51175101)on this paper.
文摘Combining deep-learning image inpainting algorithms with the microfluidic technology,the paper proposes a method to achieve dynamic stealth and camouflage by using a microfluidic vision camouflage system simulating the chameleon skin.The basic principle is to perceive color changes in the external environment and collect ambient image information,and then utilize the image inpainting algorithm to adjust the control signals of the microfluidic system in real time.The detailed working principle of the microfluidic vision camouflage system is presented,and the mechanism of generating control signals for the system through deep-learning image inpainting algorithms and image-processing techniques is elucidated.The camouflage effect of the chameleon skin is analyzed and evaluated using color similarity.Results indicate that the camouflaged images are consistent with the background environment,thereby improving the target’s stealth and maneuvering characteristics.The camouflage technology developed in the paper based on the microfluidic vision camouflage system can be applied to several situations,such as military camouflage uniforms,robot skins,and weapon equipment.
基金supported by the National Natural Science Foundation of China(Nos.52003121,2220081350 and 22301111).
文摘This study serves as a guide to the development of a polydimethylsiloxane(PDMS)-encapsulated liquid metal-MXene aerogel,which has proven to be highly effective for electromagnetic wave absorption,particularly in saline environments.Through directional freezing and casting techniques,we have optimized the sample to exhibit enhanced absorption properties,achieving a reflection loss peak of-63.10 dB at 14.36 GHz.Variations in liquid metal content were found to significantly impact the complex permittivity of the aerogel,resulting in decreases observed in both real and imaginary components.This underscores the crucial role of conductivity in electromagnetic wave damping.Simultaneously,increases in tangent loss and attenuation constant highlight the vital contribution of MXene towards dissipating electromagnetic energy.Our best sample exhibits enhanced mechanical robustness,as evidenced by a high tensile modulus of 1 MPa.Notably,this exceptional performance is sustained for an extended period of 4 weeks even under harsh conditions such as high temperature,acid mist exposure,alkaline exposure,and immersion in synthetic seawater.By testing the thermal camouflage performance,samples achieved processable and efficient camouflage performance at multiple temperatures.This comprehensive dataset confirms the adaptability of the PDMS-encapsulated liquid metal-MXene aerogel as an effective solution for electromagnetic wave absorption in challenging environmental scenarios.
基金the National Key Research and Development Program of China(No.2022ZD0210500)the National Natural Science Foundation of China(Nos.61972067,U21A20491,and 62103437)the Dalian Outstanding Youth Science Foundation(No.2022RJ01)。
文摘Deep neural networks,especially face recognition models,have been shown to be vulnerable to adversarial examples.However,existing attack methods for face recognition systems either cannot attack black-box models,are not universal,have cumbersome deployment processes,or lack camouflage and are easily detected by the human eye.In this paper,we propose an adversarial pattern generation method for face recognition and achieve universal black-box attacks by pasting the pattern on the frame of goggles.To achieve visual camouflage,we use a generative adversarial network(GAN).The scale of the generative network of GAN is increased to balance the performance conflict between concealment and adversarial behavior,the perceptual loss function based on VGG19 is used to constrain the color style and enhance GAN’s learning ability,and the fine-grained meta-learning adversarial attack strategy is used to carry out black-box attacks.Sufficient visualization results demonstrate that compared with existing methods,the proposed method can generate samples with camouflage and adversarial characteristics.Meanwhile,extensive quantitative experiments show that the generated samples have a high attack success rate against black-box models.
