Search Results (17)

This study explores how contemporary text-to-image models interpret and generate Classical and Baroque styles under Wölfflin’s framework—two categories that are atemporal and transversal across media. Our goal is to see whether generative AI can replicate the nuanced stylistic cues that art historians attribute to them. We prompted two popular models (DALL•E and Midjourney) using explicit style labels (e.g., “baroque” and “classical”) as well as more implicit cues (e.g., “dynamic”, “static”, or reworked Wölfflin descriptors). We then collected expert ratings and conducted broader qualitative reviews to assess how each output aligned with Wölfflin’s characteristics. Our findings suggest that the term “baroque” usually evokes features recognizable in typically historical Baroque artworks, while “classical” often yields less distinct results, particularly when a specified genre (portrait, still life) imposes a centered, closed-form composition. Removing explicit style labels may produce highly abstract images, revealing that Wölfflin’s descriptors alone may be insufficient to convey Classical or Baroque styles efficiently. Interestingly, the term “dynamic” gives rather chaotic images, yet this chaos is somehow ordered, centered, and has an almost Classical feel. Altogether, these observations highlight the complexity of bridging canonical stylistic frameworks and contemporary AI training biases, underscoring the need to update or refine Wölfflin’s atemporal categories to accommodate how generative models—and modern visual culture—reinterpret Classical and Baroque. Full article

Aiming at the problems of misdetection, leakage, and low recognition accuracy caused by numerous surface defects and complex backgrounds of laser nozzles, this paper proposes DEC-YOLO, a novel detection model centered on the DEC Module (DenseNet-explicit visual center composite module). The DEC Module, as the core innovation, combines the dense connectivity of DenseNet with the local–global feature integration capability of the explicit visual center (EVC) to enhance gradient propagation stability during the training process and enhance fundamental defect feature extraction. To further optimize detection performance, three auxiliary strategies are introduced: (1) a head decoupling strategy to separate classification and regression tasks, (2) cross-layer connections for multi-scale feature fusion, and (3) coordinate attention to suppress background interference. The experimental results on a custom dataset demonstrate that DEC-YOLO achieves a mean average precision (mAP@0.5) of 87.5%, surpassing that of YOLOv7 by 10.5%, and meets the accuracy and speed requirements needed in the laser cutting production environment. Full article

To address the issues of significant object discrepancies, low similarity, and image noise interference between source and target domains in object detection, we propose a supervised learning approach combined with knowledge distillation. Initially, student and teacher models are jointly trained through supervised and distillation-based approaches, iteratively refining the inter-model weights to mitigate the issue of model overfitting. Secondly, a combined convolutional module is integrated into the feature extraction network of the student model, to minimize redundant computational effort; an explicit visual center module is embedded within the feature pyramid network, to bolster feature representation; and a spatial grouping enhancement module is incorporated into the region proposal network, to mitigate the adverse effects of noise on the outcomes. Ultimately, the model undergoes a comprehensive optimization process that leverages the loss functions originating from both the supervised and knowledge distillation phases. The experimental results demonstrate that this strategy significantly boosts classification and identification accuracy on cross-domain datasets; when compared to the TFA (Task-agnostic Fine-tuning and Adapter), CD-FSOD (Cross-Domain Few-Shot Object Detection) and DeFRCN (Decoupled Faster R-CNN for Few-Shot Object Detection), with sample orders of magnitude 1 and 5, increased the detection accuracy by 1.67% and 1.87%, respectively. Full article

In recent years, achieving high-precision and high-speed target detection of surface defects on aluminum profiles to meet the requirements of industrial applications has been challenging. In this paper, the GEB-YOLO is proposed based on the YOLOv7 algorithm. First, the global attention mechanism (GAM) is introduced, highlighting defect features. Second, the Explicit Visual Center Block (EVCBlock) is integrated into the network for key information extraction. Meanwhile, the BiFPN network structure is adopted to enhance feature fusion. The ablation experiments have demonstrated that the defect detection accuracy of the GEB-YOLO model is improved by 6.3%, and the speed is increased by 15% compared to the YOLOv7 model. Full article

To address the current steel surface defect detection algorithms in practical applications involving low detection accuracy, an efficient and highly accurate strip steel surface defect detection algorithm, DEW-YOLO, is proposed in this paper. Firstly, by combining the advantages of deformable convolutional networks (DCNs), this paper innovates the C2F module in YOLOv8 and proposes a C2f_DCN module that can flexibly sample features to enhance the abilities of learning and expressing defect features of different sizes and shapes. Secondly, the explicit visual center (EVC) is introduced into the backbone network, which enhances feature extraction capabilities and adaptability and enables the model to better adjust features at different levels and scales. Finally, the original loss function is replaced with the Wise-IoU (WIoU) loss function to accurately measure the similarity between the target frames and improve the defect detection performance of the model. The experimental results on the NEU-DET dataset demonstrate that the algorithms proposed in this paper achieved a mean average precision (mAP) of 80.3% in steel surface defect detection tasks, which was a 3.9% improvement over the original YOLOv8 model. The model’s inference speed reached 91 frames per second (FPS). DEW-YOLO effectively enhances the accuracy of steel defect detection and better satisfies industrial inspection requirements. Full article

In an unmanned orchard, various tasks such as seeding, irrigation, health monitoring, and harvesting of crops are carried out by unmanned vehicles. These vehicles need to be able to distinguish which objects are fruit trees and which are not, rather than relying on human guidance. To address this need, this study proposes an efficient and robust method for fruit tree detection in orchard point cloud maps. Feature extraction is performed on the 3D point cloud to form a two-dimensional feature vector containing three-dimensional information of the point cloud and the tree target is detected through the customized deep learning network. The impact of various feature extraction methods such as average height, density, PCA, VFH, and CVFH on the detection accuracy of the network is compared in this study. The most effective feature extraction method for the detection of tree point cloud objects is determined. The ECA attention module and the EVC feature pyramid structure are introduced into the YOLOv8 network. The experimental results show that the deep learning network improves the precision, recall, and mean average precision by 1.5%, 0.9%, and 1.2%, respectively. The proposed framework is deployed in unmanned orchards for field testing. The experimental results demonstrate that the framework can accurately identify tree targets in orchard point cloud maps, meeting the requirements for constructing semantic orchard maps. Full article

Accurate comprehension of others’ thoughts and intentions is crucial for smooth social interactions, wherein understanding their perceptual experiences serves as a fundamental basis for this high-level social cognition. However, previous research has predominantly focused on the visual modality when investigating perceptual processing from others’ perspectives, leaving the exploration of multisensory inputs during this process largely unexplored. By incorporating auditory stimuli into visual perspective-taking (VPT) tasks, we have designed a novel experimental paradigm in which the spatial correspondence between visual and auditory stimuli was limited to the altercentric rather than the egocentric reference frame. Overall, we found that when individuals engaged in explicit or implicit VPT to process visual stimuli from an avatar’s viewpoint, the concomitantly presented auditory stimuli were also processed within this avatar-centered reference frame, revealing altercentric cross-modal interactions. Full article

DNA synthesis catalyzed by DNA polymerase is essential for all life forms, and phosphodiester bond formation with phosphorus center inversion is a key step in this process. Herein, by using a single-selenium-atom-modified dNTP probe, we report a novel strategy to visualize the reaction stereochemistry and catalysis. We capture the before- and after-reaction states and provide explicit evidence of the center inversion and in-line attacking S_N2 mechanism of DNA polymerization, while solving the diastereomer absolute configurations. Further, our kinetic and thermodynamic studies demonstrate that in the presence of Mg²⁺ ions (or Mn²⁺), the binding affinity (K_m) and reaction selectivity (k_cat/K_m) of dGTPαSe-Rp were 51.1-fold (or 19.5-fold) stronger and 21.8-fold (or 11.3-fold) higher than those of dGTPαSe-Sp, respectively, indicating that the diastereomeric Se-Sp atom was quite disruptive of the binding and catalysis. Our findings reveal that the third metal ion is much more critical than the other two metal ions in both substrate recognition and bond formation, providing insights into how to better design the polymerase inhibitors and discover the therapeutics. Full article

Traffic sign detection is an important research direction in the process of intelligent transportation in the Internet era, and plays a crucial role in ensuring traffic safety. The purpose of this research is to propose a traffic-sign-detection algorithm based on the selective kernel attention (SK attention), explicit visual center (EVC), and YOLOv5 model to address the problems of small targets, incomplete detection, and insufficient detection accuracy in natural and complex road situations. First, the feature map with a smaller receptive field in the backbone network is fused with other scale feature maps to increase the small target detection layer. Then, the SK attention mechanism is introduced to extract and weigh features at different scales and levels, enhancing the attention to the target. By fusing the explicit visual center to gather local area features within the layer, the detection effect of small targets is improved. According to the experiment results, the mean average precision (mAP) on the Tsinghua-Tencent Traffic Sign Dataset (TT100K) for the proposed algorithm is 88.5%, which is 4.6% higher than the original model, demonstrating the practicality of the detection of small traffic signs. Full article

Kappa-angle calibration shows its importance in gaze tracking due to the special structure of the eyeball. In a 3D gaze-tracking system, after the optical axis of the eyeball is reconstructed, the kappa angle is needed to convert the optical axis of the eyeball to the real gaze direction. At present, most of the kappa-angle-calibration methods use explicit user calibration. Before eye-gaze tracking, the user needs to look at some pre-defined calibration points on the screen, thereby providing some corresponding optical and visual axes of the eyeball with which to calculate the kappa angle. Especially when multi-point user calibration is required, the calibration process is relatively complicated. In this paper, a method that can automatically calibrate the kappa angle during screen browsing is proposed. Based on the 3D corneal centers and optical axes of both eyes, the optimal objective function of the kappa angle is established according to the coplanar constraint of the visual axes of the left and right eyes, and the differential evolution algorithm is used to iterate through kappa angles according to the theoretical angular constraint of the kappa angle. The experiments show that the proposed method can make the gaze accuracy reach 1.3${}^{°}$ in the horizontal plane and 1.34${}^{°}$ in the vertical plane, both of which are within the acceptable margins of gaze-estimation error. The demonstration of explicit kappa-angle calibration is of great significance to the realization of the instant use of gaze-tracking systems. Full article

Recent deep models trained on large-scale RGB datasets lead to considerable achievements in visual detection tasks. However, the training examples are often limited for an infrared detection task, which may deteriorate the performance of deep detectors. In this paper, we propose a transfer approach, Source Model Guidance (SMG), where we leverage a high-capacity RGB detection model as the guidance to supervise the training process of an infrared detection network. In SMG, the foreground soft label generated from the RGB model is introduced as source knowledge to provide guidance for cross-domain transfer. Additionally, we design a Background Suppression Module in the infrared network to receive the knowledge and enhance the foreground features. SMG is easily plugged into any modern detection framework, and we show two explicit instantiations of it, SMG-C and SMG-Y, based on CenterNet and YOLOv3, respectively. Extensive experiments on different benchmarks show that both SMG-C and SMG-Y achieve remarkable performance even if the training set is scarce. Compared to advanced detectors on public FLIR, SMG-Y with 77.0% mAP outperforms others in accuracy, and SMG-C achieves real-time detection at a speed of 107 FPS. More importantly, SMG-Y trained on a quarter of the thermal dataset obtains 74.5% mAP, surpassing most state-of-the-art detectors with full FLIR as training data. Full article

Despite the breakthroughs in accuracy and efficiency of object detection using deep neural networks, the performance of small object detection is far from satisfactory. Gaze estimation has developed significantly due to the development of visual sensors. Combining object detection with gaze estimation can significantly improve the performance of small object detection. This paper presents a centered multi-task generative adversarial network (CMTGAN), which combines small object detection and gaze estimation. To achieve this, we propose a generative adversarial network (GAN) capable of image super-resolution and two-stage small object detection. We exploit a generator in CMTGAN for image super-resolution and a discriminator for object detection. We introduce an artificial texture loss into the generator to retain the original feature of small objects. We also use a centered mask in the generator to make the network focus on the central part of images where small objects are more likely to appear in our method. We propose a discriminator with detection loss for two-stage small object detection, which can be adapted to other GANs for object detection. Compared with existing interpolation methods, the super-resolution images generated by CMTGAN are more explicit and contain more information. Experiments show that our method exhibits a better detection performance than mainstream methods. Full article

Germinal centers (GCs) are specialized compartments within the secondary lymphoid organs where B cells proliferate, differentiate, and mutate their antibody genes in response to the presence of foreign antigens. Through the GC lifespan, interclonal competition between B cells leads to increased affinity of the B cell receptors for antigens accompanied by a loss of clonal diversity, although the mechanisms underlying clonal dynamics are not completely understood. We present here a multi-scale quantitative model of the GC reaction that integrates an intracellular component, accounting for the genetic events that shape B cell differentiation, and an extracellular stochastic component, which accounts for the random cellular interactions within the GC. In addition, B cell receptors are represented as sequences of nucleotides that mature and diversify through somatic hypermutations. We exploit extensive experimental characterizations of the GC dynamics to parameterize our model, and visualize affinity maturation by means of evolutionary phylogenetic trees. Our explicit modeling of B cell maturation enables us to characterise the evolutionary processes and competition at the heart of the GC dynamics, and explains the emergence of clonal dominance as a result of initially small stochastic advantages in the affinity to antigen. Interestingly, a subset of the GC undergoes massive expansion of higher-affinity B cell variants (clonal bursts), leading to a loss of clonal diversity at a significantly faster rate than in GCs that do not exhibit clonal dominance. Our work contributes towards an in silico vaccine design, and has implications for the better understanding of the mechanisms underlying autoimmune disease and GC-derived lymphomas. Full article

Scene classification of high-resolution remote sensing images (HRRSI) is one of the most important means of land-cover classification. Deep learning techniques, especially the convolutional neural network (CNN) have been widely applied to the scene classification of HRRSI due to the advancement of graphic processing units (GPU). However, they tend to extract features from the whole images rather than discriminative regions. The visual attention mechanism can force the CNN to focus on discriminative regions, but it may suffer from the influence of intra-class diversity and repeated texture. Motivated by these problems, we propose an attention-based deep feature fusion (ADFF) framework that constitutes three parts, namely attention maps generated by Gradient-weighted Class Activation Mapping (Grad-CAM), a multiplicative fusion of deep features and the center-based cross-entropy loss function. First of all, we propose to make attention maps generated by Grad-CAM as an explicit input in order to force the network to concentrate on discriminative regions. Then, deep features derived from original images and attention maps are proposed to be fused by multiplicative fusion in order to consider both improved abilities to distinguish scenes of repeated texture and the salient regions. Finally, the center-based cross-entropy loss function that utilizes both the cross-entropy loss and center loss function is proposed to backpropagate fused features so as to reduce the effect of intra-class diversity on feature representations. The proposed ADFF architecture is tested on three benchmark datasets to show its performance in scene classification. The experiments confirm that the proposed method outperforms most competitive scene classification methods with an average overall accuracy of 94% under different training ratios. Full article

When performing manual actions, eye movements precede hand movements to target locations: Before we grasp an object, we look at it. Eye-hand guidance is even preserved when visual targets are unavailable, e.g., grasping behind an occlusion. This “looking-at-nothing” behavior might be functional, e.g., as “deictic pointer” for manual control or as memory-retrieval cue, or a by-product of automatization. Here, it is studied if looking at empty locations before acting on them is beneficial for sensorimotor performance. In five experiments, participants completed a click sequence on eight visual targets for 0–100 trials while they had either to fixate on the screen center or could move their eyes freely. During 50–100 consecutive trials, participants clicked the same sequence on a blank screen with free or fixed gaze. During both phases, participants looked at target locations when gaze shifts were allowed. With visual targets, target fixations led to faster, more precise clicking, fewer errors, and sparser cursor-paths than central fixation. Without visual information, a tiny free-gaze benefit could sometimes be observed and was rather a memory than a motor-calculation benefit. Interestingly, central fixation during learning forced early explicit encoding causing a strong benefit for acting on remembered targets later, independent of whether eyes moved then. Full article