Synthetic images aid the recognition of human-made art forgeries
Ontology highlight
ABSTRACT:
SUBMITTER: Ostmeyer J
PROVIDER: S-EPMC10866502 | biostudies-literature | 2024 Jan
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:The polymerization of 3,4-dihydroxy-L-phenylalanine leads to a carboxylic acid-rich synthetic melanin-like material (poly-L-DOPA). Synthetic melanin most resembles natural eumelanin in chemical structure. However, its deposition on surfaces leading to colored surfaces by interference is not as easy to accomplish as in the case of the preparation of colored surfaces by dopamine hydrochloride polymerization. This study deals with the preparation of new colored surfaces made from poly-L-DOPA displaying vivid colors by interference. These surfaces were obtained by depositing thin films of poly-L-DOPA on a reflective silicon nitride substrate. A high ionic strength in the polymerization medium was essential to accomplish the coating. The effect of ionic strength on the resulting surfaces was studied via reflectance, Atomic Force Microscopy (AFM) and Scanning Electron Microscopy (SEM). The refractive index was determined by ellipsometry, and was nearly constant to 1.8 when λ > 650 nm. In the visible spectral region, the imaginary part of the refractive index becomes relevant. The refractive index in the visible wavelength range (400-600 nm) was in the range 1.7-1.80.
| S-EPMC8468714 | biostudies-literature
Project description:The bioengineering applications of cells, such as cell printing and multicellular assembly, are directly limited by cell damage and death due to a harsh environment. Improved cellular robustness thus motivates investigations into cell encapsulation, which provides essential protection. Here we target the cell-surface glycocalyx and cross-link two layers of DNA nanorods on the cellular plasma membrane to form a modular and programmable nanoshell. We show that the DNA origami nanoshell modulates the biophysical properties of cell membranes by enhancing the membrane stiffness and lowering the lipid fluidity. The nanoshell also serves as armor to protect cells and improve their viability against mechanical stress from osmotic imbalance, centrifugal forces, and fluid shear stress. Moreover, it enables mediated cell-cell interactions for effective and robust multicellular assembly. Our results demonstrate the potential of the nanoshell, not only as a cellular protection strategy but also as a platform for cell and cell membrane manipulation.
| S-EPMC10416349 | biostudies-literature
Project description:Most people encounter art images as digital reproductions on a computer screen instead of as originals in a museum or gallery. With the development of digital technologies, high-resolution artworks can be accessed anywhere and anytime by a large number of viewers. Since these digital images depict the same content and are attributed to the same artist as the original, it is often implicitly assumed that their aesthetic evaluation will be similar. When it comes to the digital reproductions of art, however, it is also obvious that reproductions do differ from the originals in various aspects. Besides image quality, resolution, and format, the most obvious change is in the representation of color. The effects of subjectively varying surface-level image features on art evaluation have not been clearly assessed. To address this gap, we compare the evaluation of digital reproductions of 16 expressionist and impressionist paintings manipulated to have a high color saturation vs. a saturation similar to the original. We also investigate the impact of viewing time (100 ms vs. unrestricted viewing time) and expertise (art experts vs. laypersons), two other aspects that may impact the perception of art in online contexts. Moreover, we link these dimensions to a recent model of aesthetic experience [the Vienna Integrated Model of Top-Down and Bottom-Up Processes in Art Perception (VIMAP)]. Results suggest that color saturation does not exert a major influence on liking. Cognitive and emotional aspects (interest, confusion, surprise, and boredom), however, are affected - to different extents for experts and laypersons. For laypersons, the increase in color saturation led to more positive assessments of an artwork, whereas it resulted in increased confusion for art experts. This insight is particularly important when it comes to reproducing artworks digitally. Depending on the intended use, increasing or decreasing the color saturation of the digitally reproduced image might be most appropriate. We conclude with a discussion of these findings and address the question of why empirical aesthetics requires more precise dimensions to better understand the subtle processes that take place in the perception of today's digitally reproduced art environment.
| S-EPMC7759521 | biostudies-literature
Project description:Digital image processing is a constantly evolving field encompassing a wide range of techniques and applications. Researchers worldwide are continually developing various algorithms across multiple fields to achieve accurate image classification. Advanced computer vision algorithms are crucial for architectural and artistic analysis. The digitalization of art has significantly enhanced the accessibility and conservation of fine-art paintings, yet the risk of art theft remains a significant challenge. Improving art security necessitates the precise identification of fine-art paintings. Although current recognition systems have shown potential, there is significant scope for enhancing their efficiency. We developed an improved recognition system for categorizing fine-art paintings using convolutional transformers, specified by an attention mechanism to enhance focused learning on the data. As part of the most advanced architectures in the deep learning family, transformers are empowered by a multi-head attention mechanism, thus improving learning efficiency. To assess the performance of our model, we compared it with those developed using four pre-trained networks: ResNet50, VGG16, AlexNet, and ViT. Each pre-trained network was integrated into a corresponding state-of-the-art model as the first processing blocks. These four state-of-the-art models were constructed under the transfer learning strategy, one of the most commonly used approaches in this field. The experimental results showed that our proposed system outperformed the other models. Our study also highlighted the effectiveness of using convolutional transformers for learning image features.
| S-EPMC11622946 | biostudies-literature
Project description:This paper introduces Art_GenEvalGPT, a novel dataset of synthetic dialogues centered on art generated through ChatGPT. Unlike existing datasets focused on conventional art-related tasks, Art_GenEvalGPT delves into nuanced conversations about art, encompassing a wide variety of artworks, artists, and genres, and incorporating emotional interventions, integrating speakers' subjective opinions and different roles for the conversational agents (e.g., teacher-student, expert guide, anthropic behavior or handling toxic users). Generation and evaluation stages of GenEvalGPT platform are used to create the dataset, which includes 13,870 synthetic dialogues, covering 799 distinct artworks, 378 different artists, and 26 art styles. Automatic and manual assessment proof the high quality of the synthetic dialogues generated. For the profile recovery, promising lexical and semantic metrics for objective and factual attributes are offered. For subjective attributes, the evaluation for detecting emotions or subjectivity in the interventions achieves 92% of accuracy using LLM-self assessment metrics.
| S-EPMC11283562 | biostudies-literature
Project description:When an individual participates in empirical studies involving the visual arts, they most often are presented with a stream of images, shown on a computer, depicting reproductions of artworks by respected artists but which are often not known to the viewer. While art can of course be shown in presentia actuale-e.g., in the museum-this laboratory paradigm has become our go-to basis for assessing interaction, and, often in conjunction with some means of rating, for assessing evaluative, emotional, cognitive, and even neurophysiological response. However, the question is rarely asked: Do participants actually believe that every image that they are viewing is indeed "Art"? Relatedly, how does this evaluation relate to aesthetic appreciation, and do the answers to these questions vary in accordance with different strategies and interpersonal differences? In this paper, we consider the spontaneous classification of digital reproductions as art or not art. Participants viewed a range of image types-Abstract, Hyperrealistic, Poorly Executed paintings, Readymade sculptures, as well as Renaissance and Baroque paintings. They classified these as "art" or "not art" using both binary and analog scales, and also assessed for liking. Almost universally, individuals did not find all items within a class to be "art," nor did all participants agree on the arthood status for any one item. Art classification in turn showed a significant positive correlation with liking. Whether an object was classified as art moreover correlated with specific personality variables, tastes, and decision strategies. The impact of these findings is discussed for selection/assessment of participants and for better understanding the basis of findings in past and future empirical art research.
| S-EPMC5640778 | biostudies-literature
Project description:Functional MRI (fMRI) data acquired using echo-planar imaging (EPI) are highly distorted by magnetic field inhomogeneities. Distortion and differences in image contrast between EPI and T1-weighted and T2-weighted (T1w/T2w) images makes their alignment a challenge. Typically, field map data are used to correct EPI distortions. Alignments achieved with field maps can vary greatly and depends on the quality of field map data. However, many public datasets lack field map data entirely. Additionally, reliable field map data is often difficult to acquire in high-motion pediatric or developmental cohorts. To address this, we developed Synth, a software package for distortion correction and cross-modal image registration that does not require field map data. Synth combines information from T1w and T2w anatomical images to construct an idealized undistorted synthetic image with similar contrast properties to EPI data. This synthetic image acts as an effective reference for individual-specific distortion correction. Using pediatric (ABCD: Adolescent Brain Cognitive Development) and adult (MSC: Midnight Scan Club; HCP: Human Connectome Project) data, we demonstrate that Synth performs comparably to field map distortion correction approaches, and often outperforms them. Field map-less distortion correction with Synth allows accurate and precise registration of fMRI data with missing or corrupted field map information.
| S-EPMC10106483 | biostudies-literature
Project description:BackgroundArtificial intelligence (AI) in medical imaging diagnostics has huge potential, but human judgement is still indispensable. We propose an AI-aided teaching method that leverages generative AI to train students on many images while preserving patient privacy.MethodsA web-based course was designed using 600 synthetic ultra-widefield (UWF) retinal images to teach students to detect disease in these images. The images were generated by stable diffusion, a large generative foundation model, which we fine-tuned with 6285 real UWF images from six categories: five retinal diseases (age-related macular degeneration, glaucoma, diabetic retinopathy, retinal detachment and retinal vein occlusion) and normal. 161 trainee orthoptists took the course. They were evaluated with two tests: one consisting of UWF images and another of standard field (SF) images, which the students had not encountered in the course. Both tests contained 120 real patient images, 20 per category. The students took both tests once before and after training, with a cool-off period in between.ResultsOn average, students completed the course in 53 min, significantly improving their diagnostic accuracy. For UWF images, student accuracy increased from 43.6% to 74.1% (p<0.0001 by paired t-test), nearly matching the previously published state-of-the-art AI model's accuracy of 73.3%. For SF images, student accuracy rose from 42.7% to 68.7% (p<0.0001), surpassing the state-of-the-art AI model's 40%.ConclusionSynthetic images can be used effectively in medical education. We also found that humans are more robust to novel situations than AI models, thus showcasing human judgement's essential role in medical diagnosis.
| S-EPMC11503156 | biostudies-literature
Project description:We recently found that luminance edges are more evenly distributed across orientations in large subsets of traditional artworks, i.e., artworks are characterized by a relatively high entropy of edge orientations, when compared to several categories of other (non-art) images. In the present study, we asked whether edge-orientation entropy is associated with aesthetic preference in a wide variety of other man-made visual patterns and scenes. In the first (exploratory) part of the study, participants rated the aesthetic appeal of simple shapes, artificial ornamental patterns, facades of buildings, scenes of interior architecture, and music album covers. Results indicated that edge-orientation entropy predicts aesthetic ratings for these stimuli. However, the magnitude of the effect depended on the type of images analyzed, on the range of entropy values encountered, and on the type of aesthetic rating (pleasing, interesting, or harmonious). For example, edge-orientation entropy predicted about half of the variance when participants rated facade photographs for pleasing and interesting, but only for 3.5% of the variance for harmonious ratings of music album covers. We also asked whether edge-orientation entropy relates to the well-established human preference for curved over angular shapes. Our analysis revealed that edge-orientation entropy was as good or an even better predictor for the aesthetic ratings than curvilinearity. Moreover, entropy could substitute for shape, at least in part, to predict the aesthetic ratings. In the second (experimental) part of this study, we generated complex line stimuli that systematically varied in their edge-orientation entropy and curved/angular shape. Here, edge-orientation entropy was a more powerful predictor for ratings of pleasing and harmonious than curvilinearity, and as good a predictor for interesting. Again, the two image properties shared a large portion of variance between them. In summary, our results indicate that edge-orientation entropy predicts aesthetic ratings in diverse man-made visual stimuli. Moreover, the preference for high edge-orientation entropy shares a large portion of predicted variance with the preference for curved over angular stimuli.
| S-EPMC6172329 | biostudies-literature
Project description:Gesture recognition in non-intrusive muscle-computer interfaces is usually based on windowed descriptive and discriminatory surface electromyography (sEMG) features because the recorded amplitude of a myoelectric signal may rapidly fluctuate between voltages above and below zero. Here, we present that the patterns inside the instantaneous values of high-density sEMG enables gesture recognition to be performed merely with sEMG signals at a specific instant. We introduce the concept of an sEMG image spatially composed from high-density sEMG and verify our findings from a computational perspective with experiments on gesture recognition based on sEMG images with a classification scheme of a deep convolutional network. Without any windowed features, the resultant recognition accuracy of an 8-gesture within-subject test reached 89.3% on a single frame of sEMG image and reached 99.0% using simple majority voting over 40 frames with a 1,000 Hz sampling rate. Experiments on the recognition of 52 gestures of NinaPro database and 27 gestures of CSL-HDEMG database also validated that our approach outperforms state-of-the-arts methods. Our findings are a starting point for the development of more fluid and natural muscle-computer interfaces with very little observational latency. For example, active prostheses and exoskeletons based on high-density electrodes could be controlled with instantaneous responses.
| S-EPMC5109222 | biostudies-literature