Attentional Guidance Research Articles

Dual-Metric Neural Network With Attention Guidance for Surface Defect Few-Shot Detection in Smart Manufacturing

Abstract As an important application of human–robot collaboration, intelligent detection of surface defects is crucial for production quality control, which also helps in relieving the workload of technical staff in human-centric smart manufacturing. To accurately detect defects with limited samples in industrial practice, a dual-metric neural network with attention-guided is proposed. First, an attention-guided recognition network with channel attention and position attention module is designed to efficiently learn representative defect features with limited samples. Second, aiming to detect defects with confusing surface images, a dual-metric function is presented to learn the classification boundary by controlling the distance of samples in feature space from intraclass and interclass. The experiment results on the fabric defect dataset demonstrate that the proposed approach outperforms state-of-the-art methods in accuracy, recall, precision, F1-score, and few-shot accuracy. Further comparative experiments reveal that the dual-metric function is superior in improving the few-shot detection accuracy.

The dynamics of attentional guidance by working memory contents

Working memory (WM) contents can guide attention toward matching sensory information in the environment, but there are mixed findings regarding whether only a single prioritized item or multiple items held in WM can guide attention. The present study examines the limit of WM-guided attention with a novel task procedure and mouse trajectory analysis. Specifically, we introduced a perceptual-matching task utilizing the continuous estimation procedure within the maintenance interval of a WM task for one or two colors. We found that the overall perceptual matching mouse trajectory were robustly biased toward the location of WM-match color on the color-wheel (i.e., attraction bias), but only at memory set size one. However, the analysis of circular mouse trajectory distributions, through hierarchical Bayesian modeling, revealed two separable central peaks at both memory set sizes. Furthermore, model-free analysis demonstrated that the perceptual matching mouse trajectory patterns were similar regardless of memory set sizes. Together, these results support the single-item account and highlight the utility of mouse trajectory analyses in hypothesis testing in experimental psychology.

RTAU-Net: A novel 3D rectal tumor segmentation model based on dual path fusion and attentional guidance

Background and objectiveAccording to the Global Cancer Statistics 2020, colorectal cancer has the third-highest diagnosis rate (10.0 %) and the second-highest mortality rate (9.4 %) among the 36 types. Rectal cancer accounts for a large proportion of colorectal cancer. The size and shape of the rectal tumor can directly affect the diagnosis and treatment by doctors. The existing rectal tumor segmentation methods are based on two-dimensional slices, which cannot analyze a patient's tumor as a whole and lose the correlation between slices of MRI image, so the practical application value is not high. MethodsIn this paper, a three-dimensional rectal tumor segmentation model is proposed. Firstly, image preprocessing is performed to reduce the effect caused by the unbalanced proportion of background region and target region, and improve the quality of the image. Secondly, a dual-path fusion network is designed to extract both global features and local detail features of rectal tumors. The network includes two encoders, a residual encoder for enhancing the spatial detail information and feature representation of the tumor and a transformer encoder for extracting global contour information of the tumor. In the decoding stage, we merge the information extracted from the dual paths and decode them. In addition, for the problem of the complex morphology and different sizes of rectal tumors, a multi-scale fusion channel attention mechanism is designed, which can capture important contextual information of different scales. Finally, visualize the 3D rectal tumor segmentation results. ResultsThe RTAU-Net is evaluated on the data set provided by Shanxi Provincial Cancer Hospital and Xinhua Hospital. The experimental results showed that the Dice of tumor segmentation reached 0.7978 and 0.6792, respectively, which improved by 2.78 % and 7.02 % compared with suboptimal model. ConclusionsAlthough the morphology of rectal tumors varies, RTAU-Net can precisely localize rectal tumors and learn the contour and details of tumors, which can relieve physicians' workload and improve diagnostic accuracy.

Transposition, not translation: Recuperating attentionality on Pantelleria, Sicily

AbstractThis article examines how elderly rural Sicilians recall the meanings of words rendered obsolete by infrastructural, technological, and economic changes that occurred in their lifetime. I examine conversations from my 2016 and 2019 fieldwork on Pantelleria, Sicily, characterized by what I term recuperated attentionality, speaking from erstwhile attentional circumstances. To unpack the meanings of words, elderly islanders employ transposition, contextualizing their attentional guidance from a moment of reference anchored in the remembered past, orienting to an obsolete way of being in the world. Socio‐biographical discontinuity means that acquaintance and familiarity with the denotata of these words is asymmetrical, once accessed by participating in island life, issuing and responding to directives, attending to tasks, and so on, but now accessed principally by memory. I examine conversational discourse in which transposition is used to unpack word meanings, which clashes with elicitation norms that request translational equivalents for words.

ICA-Net: Industrial defect detection network based on convolutional attention guidance and aggregation of multiscale features

Detecting surface defects in the industry is essential for improving the quality of industrial products and maintaining product safety. However, problems such as the similarity of defects, significant variation in the scale of the target object, and the balance between detection speed and accuracy in industrial inspection scenarios have been considerable research topics in this field. This paper proposes an industrial defect detection network based on convolutional attention-guided and aggregated multiscale features to address these issues (ICA-Net). Firstly, for similarity defects in complex backgrounds, this paper proposes a backbone network with a combination of lightweight convolutional blocks and self-attentive modules to fully extract images’ local and global information and enhance the network’s expressiveness. Secondly, to make full use of the shallow fine-grained features and deep semantic features of the backbone network to improve the detection capability of defects with significant scale changes, this paper designs a cross-layer multiscale feature fusion network (CEF-Net), which fully fuses the features of adjacent layers and cross-layers through a reweighting feature strategy to enrich the network feature transfer path and ensure the efficient fusion of different scale features in the network. At the same time, the fine-grained feature fusion module (FFM) is used to fuse elements from multiple layers to extract more contextual information, enhance the extraction of fine-grained features and improve the detection capability of complex small targets. Finally, to address the problems of inaccurate regression localization and low detection accuracy of defects in existing industrial algorithms, a new IoU loss function (G-IOU) is proposed for regressing the intersection part of the predicted frame and the actual structure according to the aspect ratio of the real frame during the model regression to improve the accuracy and stability of detection. The experimental results show that 94.1%, 98.6%, 99.4%, 98.8% and 96.5% of mAP@.5 are obtained on steel, PCB, aluminium, automobile and Xsteel steel metal surface defect datasets, respectively, and 48 FPS is achieved, which is superior to the current mainstream detectors and meets the needs of practical industrial production.

Hybrid Dilated Convolution with Attention Mechanisms for Image Denoising

In the field of image denoising, convolutional neural networks (CNNs) have become increasingly popular due to their ability to learn effective feature representations from large amounts of data. In the field of image denoising, CNNs are widely used to improve performance. However, increasing network depth can weaken the influence of shallow layers on deep layers, especially for complex denoising tasks such as real denoising and blind denoising, where conventional networks fail to achieve high-quality results. To address this issue, this paper proposes a hybrid dilated convolution-based denoising network (AMDNet) that incorporates attention mechanisms. In specific, AMDNet consists of four modules: the sparse module (SM), the feature fusion module (FFM), the attention guidance module (AGM), and the image residual module (IRM). The SM employs hybrid dilated convolution to extract local features, while the FFM is used to integrate global and local features. The AGM accurately extracts noise information hidden in complex backgrounds. Finally, the IRM reconstructs images in a residual manner to obtain high-quality results after denoising. AMDNet has the following features: (1) The sparse mechanism in hybrid dilated convolution enables better extraction of local features, enhancing the network’s ability to capture noise information. (2) The feature fusion module, through long-range connections, fully integrates global and local features, improving the performance of the model; (3) the attention module is ingeniously designed to precisely extract features in complex backgrounds. The experimental results demonstrate that AMDNet achieves outstanding performance on three tasks (Gaussian noise, real noise, and blind denoising).

Tracking semantic relatedness: numeral classifiers guide gaze to visual world objects

Directing visual attention toward items mentioned within utterances can optimize understanding the unfolding spoken language and preparing appropriate behaviors. In several languages, numeral classifiers specify semantic classes of nouns but can also function as reference trackers. Whereas all classifier types function to single out objects for reference in the real world and may assist attentional guidance, we propose that only sortal classifiers efficiently guide visual attention by being inherently attached to the nouns' semantics, since container classifiers are pragmatically attached to the nouns they classify, and the default classifiers index a noun without specifying the semantics. By contrast, container classifiers are pragmatically attached, and default classifiers index a noun without specifying the semantics. Using eye tracking and the “visual world paradigm”, we had Chinese speakers (N = 20) listen to sentences and we observed that they looked spontaneously within 150 ms after offset of the Sortal classifier. After about 200 ms the same occurred for the container classifiers, but with the default classifier only after about 700 ms. This looking pattern was absent in a control group of non-Chinese speakers and the Chinese speakers' gaze behavior can therefore only be ascribed to classifier semantics and not to artifacts of the visual objects. Thus, we found that classifier types affect the rapidity of spontaneously looking at the target objects on a screen. These significantly different latencies indicate that the stronger the semantic relatedness between a classifier and its noun, the more efficient the deployment of overt attention.

Symbolic purposes of style: Ruben Östlund’s The Square

Symbolic criticism is easily reduced to an act of translating meaning, remote from the style of rich narratives. Drawing on David Bordwell’s distinction between broad, overlapping functions of style, I argue that we can examine symbolic purposes as secondary to decorative and expressive purposes. I show how a theory on conceptual cognition proposed by psychologist Lawrence Barsalou can let the critic transcend firm distinctions between abstract and concrete features of a narrative. This proves especially helpful when narrative purposes are downplayed, for example in the resonant elements of art cinema, affecting how we construct and experience the storylines. I posit a framework for the analysis of symbolic purposes based on general mechanisms of memory and attentional guidance, letting us see continuity with critics such as Bordwell and Victor F. Perkins. This approach enables the critic to distinguish between degrees of symbolic power in Ruben Östlund’s The Square.

Proactive suppression can be applied to multiple salient distractors in visual search.

There is a growing body of research demonstrating that the capture of attention by a single salient distractor can be prevented via proactive suppression. In real-world contexts, there are often several distracting events that compete for attention, but it is entirely unknown whether multiple objects can be suppressed concurrently. We used behavioral and electrophysiological measures to investigate the existence and time course of multiple-item suppression. We employed search displays that contained either one or two uniquely colored distractors that differed in their salience (S+ and S-), or no such distractors. Search performance improved with the number of salient distractors, indicating that the suppression of multiple items reduced the effective display set size. This was also the case when the target color was no longer fully predictable, ruling out an alternative explanation in terms of attentional guidance by target templates. In an experiment where S+ and S- always appeared together in the same display, the PD component (a marker of proactive suppression) was triggered exclusively by the more salient distractor (S+), indicative of single-item suppression. However, when displays with one or both salient distractors were intermixed, a reliable PD component was also triggered by S-, even when it was accompanied by S+ in the same display. These results show that multiple concurrent salient signals can be proactively inhibited. They demonstrate that signal suppression processes can be adaptively employed to counteract visual distraction at different locations, in order to facilitate the attentional selection of relevant objects in crowded visual environments. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

MwdpNet: towards improving the recognition accuracy of tiny targets in high-resolution remote sensing image

This study aims to develop a deep learning model to improve the accuracy of identifying tiny targets on high resolution remote sensing (HRS) images. We propose a novel multi-level weighted depth perception network, which we refer to as MwdpNet, to better capture feature information of tiny targets in HRS images. In our method, we introduce a new group residual structure, S-Darknet53, as the backbone network of our proposed MwdpNet, and propose a multi-level feature weighted fusion strategy that fully utilizes shallow feature information to improve detection performance, particularly for tiny targets. To fully describe the high-level semantic information of the image, achieving better classification performance, we design a depth perception module (DPModule). Following this step, the channel attention guidance module (CAGM) is proposed to obtain attention feature maps for each scale, enhancing the recall rate of tiny targets and generating candidate regions more efficiently. Finally, we create four datasets of tiny targets and conduct comparative experiments on them. The results demonstrate that the mean Average Precision (mAP) of our proposed MwdpNet on the four datasets achieve 87.0%, 89.2%, 78.3%, and 76.0%, respectively, outperforming nine mainstream object detection algorithms. Our proposed approach provides an effective means and strategy for detecting tiny targets on HRS images.

Multiscale Feature Aggregation Network for Salient Object Detection in Optical Remote Sensing Images

Optical remote sensing images (ORSI) have various applications in different fields, and salient target detection (ORSI-SOD) of ORSI has become an important research topic in recent years. However, ORSI-SOD is a challenging problem due to the variable and complex backgrounds, large differences in levels, mixed backgrounds, and diverse topological shapes of ORSI. In this paper, we propose a novel model called multi-scale feature refinement aggregation network (MFANet), which consists of a multi-scale feature refinement module (MFR) and a context aggregation module (CFA). The MFR module extracts semantic information of ORSI across different dimensions in the multi-scale feature extraction stage. In the feature refinement stage, we use the proposed self-refinement module under the guidance of attention and reverse attention to progressively refine the prediction results. The CFA module introduces the hybrid attention module to gradually aggregate and extract salient regions from the context extraction module. To adapt to dense scenes, we develop a hybrid loss function that enables network optimization of multi-scale objectives in a self-adaptive manner. Our method outperforms most state-of-the-art salient object detection methods proposed in recent years in terms of accuracy.

Control over attentional capture within 170 ms by long-term memory control settings: Evidence from the N2pc.

Observers adopt attentional control settings (ACSs) based on their goals that guide the capture of attention: Searched-for stimuli capture attention, and stimuli that are not searched for do not. While previous behavioural research indicates that observers can adopt long-term memory (LTM) ACSs (Giammarco et al. Visual Cognition, 24, 78-101, 2016), it seems surprising that representations in LTM could guide attention quickly enough to control attentional capture. To assess the claim that LTM ACSs exert control over early attentional orienting, we recorded electroencephalography while participants studied and searched for 30 target objects in an attention cueing task. Participants reported the studied target and ignored the preceding cues. To control for perceptual evoked responses, on each trial we presented two cue objects (one studied and one nonstudied). Even though participants were instructed to ignore the cues, studied cues produced the N2pc event-related potential, indicating early attentional orienting that was preferentially directed towards the studied cue versus the nonstudied cue. Critically, the N2pc was detectable within 170 ms, confirming that LTM ACSs rapidly control early capture. We propose an update to contemporary models of attentional capture to account for rapid attentional guidance by LTM ACSs.

Testing a relational account of search templates in visual foraging

Search templates guide human visual attention toward relevant targets. Templates are often seen as encoding exact target features, but recent studies suggest that templates rather contain “relational properties” (e.g., they facilitate “redder” stimuli instead of specific hues of red). Such relational guidance seems helpful in naturalistic searches where illumination or perspective renders exact feature values unreliable. So far relational guidance has only been demonstrated in rather artificial single-target search tasks with briefly flashed displays. Here, we investigate whether relational guidance also occurs when humans interact with the search environment for longer durations to collect multiple target elements. In a visual foraging task, participants searched for and collected multiple targets among distractors of different relationships to the target colour. Distractors whose colour differed from the environment in the same direction as the targets reduced foraging efficiency to the same amount as distractors whose colour matched the target colour. Distractors that differed by the same colour distance but in the opposite direction of the target colour did not reduce efficiency. These findings provide evidence that search templates encode relational target features in naturalistic search tasks and suggest that attention guidance based on relational features is a common mode in dynamic, real-world search environments.

Former target representations reach forward in time and proactively interfere with attentional guidance

Former target representations reach forward in time and proactively interfere with attentional guidance

Multi-task learning for the bearing based on a one-dimensional convolutional neural network with attention guidance mechanism and multi-scale feature extraction

Fault type and fault degree identification are the main aim in the bearing multi-task learning. However, a large number of on-site accidents have shown that the bearing working condition plays an important role in bearing service life and fault diagnosis. In current studies, the bearing working condition identification task is often used for auxiliary tasks and is easily ignored. Thus, this paper studies the bearing multi-task learning, which regards the working condition identification task as an equally important task. However, simply adding the working condition identification task to the frequently used multi-task model will lead to a reduction in the overall performance of the network. To solve the network performance degradation problem, a succinct and effective multi-task one-dimensional convolutional neural network with attention guidance mechanism and multi-scale feature extraction (MAM-1DCNN) is proposed. Firstly, the time-series signal is selected as the input of the MAM-1DCNN model. Secondly, the shared network of the MAM-1DCNN model applies a double-layer multi-scale convolutional neural network structure to extract more complete information. Finally, the MAM-1DCNN applies an improved attention guidance mechanism to enhance the feature application ability of different branch tasks. Through two general bearings datasets, this paper verifies the effectiveness and generalisation ability of the MAM-1DCNN model.

Feature intertrial priming biases attentional priority: Evidence from the capture-probe paradigm.

Prior experience has a strong impact on search performance, and most recent models of attention incorporate selection history as an important source of attentional guidance. Here, we focused on feature intertrial priming, a robust effect showing that responses to a singleton target are considerably faster when its unique feature repeats versus changes across successive trials. Previous studies showed that such target repetition does not reliably reduce the interference exerted by a salient distractor. This finding has been taken to indicate that target repetition does not enhance the target's competitive edge relative to the salient distractor. Therefore, it challenges the notion that intertrial priming modulates attentional priority. Here, we suggest that this inference may be misguided because the interpretation of distractor interference as a measure of the attentional priority of the salient distractor relative to the target is incorrect. To obtain a more direct measure of the impact of feature intertrial priming on the target's priority relative to a salient distractor and nontargets, we used the capture-probe paradigm. Across two experiments, probe reports from the target location increased at the expense of the salient distractor and nontarget locations when the target feature repeated versus changed, whereas distractor interference was unaffected. These findings show that feature intertrial repetition influences attentional priority. They also clearly illustrate that distractor interference indexes the priority of the salient distractor relative to the nontarget it replaces, not relative to the target-a reinterpretation that has important implications for the field of attentional capture. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Explicit attentional goals unlock implicit spatial statistical learning.

People can quickly learn spatial distributions of targets and direct attention to likely regions of targets. These implicitly learned spatial biases have been shown to be persistent, transferring to other similar visual search tasks. However, a persistent attentional bias is incompatible with frequently changing goals in our typical daily environment. We propose a flexible goal-specific probability cueing mechanism to address this discrepancy. We examined whether participants could learn and utilize target-specific spatial priority maps across five experiments (each N = 24). In Experiment 1, participants were faster to find the target at the target-specific high-probability location, in line with a goal-specific probability cueing effect. This demonstrated that separate spatial priorities derived from statistical learning can be flexibly activated based on the current goal. In Experiment 2, we ensured the results were not driven solely by intertrial priming. In Experiment 3, we ensured the results were driven by early attentional guidance effects. In Experiment 4, we extended our findings to a complex spatial distribution including four locations, supporting a sophisticated representation of target likelihood in the activated spatial priority maps. Finally, in Experiment 5, we confirmed that the effect was driven by the activation of an attentional template and not associative learning between the target cue and a spatial location. Our findings demonstrate a previously unrecognized mechanism for flexibility within statistical learning. The goal-specific probability cueing effect relies on coordination of feature-based and location-based attention, utilizing information that crosses traditional boundaries between top-down control and selection history. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Understanding the effect of a virtual moderator on people’s perception in remote discussion using social VR

Social VR enables people to join a remote discussion by keeping a high social presence and physical proximity using embodied avatars. However, the missing nonverbal cues, such as mutual eye contact and nuanced facial expression, make it challenging for distributed members to manage turn-taking, which could lead to unequal participation and affect trust building. Therefore, we propose a virtual moderator to make distributed members feel included by seeing their nonverbal behavior. The virtual moderator was designed with a “prompt Q&A″ feature to enable users to share feedback and an “attention guidance” feature to encourage participation. The preliminary result of a controlled experiment in social VR with 30 participants showed that seeing the virtual moderator’s body orientation enhanced participants’ psychological safety. In contrast, the prompt Q&A feature enhanced the perceived co-presence of their remote counterparts. We discussed how nonverbal behavior could be designed using a virtual moderator to shape human perception of the group discussion in social VR. We also pointed out challenges when providing multiple supports simultaneously in social VR.

Visual attentional guidance during smooth pursuit eye movements: Distractor interference is independent of distractor-target similarity.

In the current study, we used abrupt-onset distractors similar and dissimilar in luminance to the target of a smooth pursuit eye-movement to test if abrupt-onset distractors capture attention in a top-down or bottom-up fashion while the eyes track a moving object. Abrupt onset distractors were presented at different positions relative to the current position of a pursuit target during the closed-loop phase of smooth pursuit. Across experiments, we varied the duration of the distractors, their motion direction, and task-relevance. We found that abrupt-onset distractors decreased the gain of horizontally directed smooth-pursuit eye-movements. This effect, however, was independent of the similarity in luminance between distractor and target. In addition, distracting effects on horizontal gain were the same, regardless of the exact duration and position of the distractors, suggesting that capture was relatively unspecific and short-lived (Experiments 1 and 2). This was different with distractors moving in a vertical direction, perpendicular to the horizontally moving target. In line with past findings, these distractors caused suppression of vertical gain (Experiment 3). Finally, making distractors task-relevant by asking observers to report distractor positions increased the pursuit gain effect of the distractors. This effect was also independent of target-distractor similarity (Experiment 4). In conclusion, the results suggest that a strong location signal exerted by the pursuit targets led to very brief and largely location-unspecific interference through the abrupt onsets and that this interference was bottom-up, implying that the control of smooth pursuit was independent of other target features besides its motion signal.

Mission impossible? Spatial context relearning following a target relocation event depends on cue predictiveness

Visual search for a target is faster when the spatial layout of distractors is repeatedly encountered, illustrating that statistical learning of contextual invariances facilitates attentional guidance (contextual cueing; Chun & Jiang, 1998, Cognitive Psychology, 36, 28–71). While contextual learning is usually relatively efficient, relocating the target to an unexpected location (within an otherwise unchanged search layout) typically abolishes contextual cueing and the benefits deriving from invariant contexts recover only slowly with extensive training (Zellin et al., 2014, Psychonomic Bulletin & Review, 21(4), 1073–1079). However, a recent study by Peterson et al. (2022, Attention, Perception, & Psychophysics, 84(2), 474–489) in fact reported rather strong adaptation of spatial contextual memories following target position changes, thus contrasting with prior work. Peterson et al. argued that previous studies may have been underpowered to detect a reliable recovery of contextual cueing after the change. However, their experiments also used a specific display design that frequently presented the targets at the same locations, which might reduce the predictability of the contextual cues thereby facilitating its flexible relearning (irrespective of statistical power). The current study was a (high-powered) replication of Peterson et al., taking into account both statistical power and target overlap in context-memory adaptation. We found reliable contextual cueing for the initial target location irrespective of whether the targets shared their location across multiple displays, or not. However, contextual adaptation following a target relocation event occurred only when target locations were shared. This suggests that cue predictability modulates contextual adaptation, over and above a possible (yet negligible) influence of statistical power.