教学课程资源库 更多>>

关键词： Pedestrians   Surveys   Trajectory   Target tracking   Cameras   Reviews   Object tracking   Circuits and systems   Computer science   Computational modeling   Multiple object tracking   multiple pedestrian tracking   occlusion handling   trajectory prediction   trajectory recovery  

摘要： As an intermediate task in computer vision, multiple pedestrian tracking (MPT) aiming at tracking the pedestrians from a given video, has attracted attention due to its potential academic and commercial value. However, pedestrians commonly suffer from occlusion due to diverse and complex scenarios, which increases the challenge of this task. This survey provides comprehensive review in terms of occlusion scenarios encountered during MPT, and investigates the model robustness of the existing methods in this scenarios. Firstly, this survey introduces the various and states of occlusion. Secondly, the related occlusion datasets are introduced. Subsequently, we categorize existing occlusion handling methods according to the tracking process and detail their pros and cons. In addition, occlusion handling precision (OHP) metric is proposed to evaluate the ability of a tracker in handling occlusion in this survey. Moreover, comprehensive analyzes and discussions in several public datasets are provided to verify the effectiveness of these methods. Finally, the existing issues and future directions for occlusion handling methods are discussed. In doing so, this work serves as a foundation for future research by providing researchers with information about the occlusion handling method of MPT.

关键词： Buildings   Optimization   Collaboration   Global communication   Cultural differences   Videos   Mathematical models   History   Computer science   Chaos   Collaborative recommendation   environment-classes   agents   roles   groups   and objects (E-CARGO)   national image resource   targeted international communication  

摘要： With the acceleration of globalization, the targeted international communication of national images contributes to enhancing a nation's soft power and international recognition. It is challenging to select appropriate resources from the mass candidates for creating promotional works of national image. Existing research only focuses on the methodologies and lacks the systematic modeling and solving of national image resources recommendation. A collaborative recommendation approach to national image resources is proposed for targeted international communication. In it, the multidimensional features of national image resources and characteristics of communication audiences are modeled, and an evaluation mechanism is proposed to measure the comprehensive compatibility between national image resources and communication audiences. By innovatively introducing the role-based collaboration (RBC) theory and the environment-classes, agents, roles, groups, and objects (E-CARGO) model, the national image resources recommendation is formalized as a collaborative optimization problem. The mathematical model is built and solved via an optimization package. Finally, the case study and experiments show that the approach is efficient, feasible, and conducive to enhancing the efficiency of selecting national image resources. It offers a novel research paradigm for targeted international communication.

关键词： Testing   Complexity theory   Probability distribution   Computer science   Upper bound   Surveys   Polynomials   Machine learning algorithms   Machine learning   Limiting   Distribution testing   tolerant testing   non-tolerant testing   sample complexity  

摘要： The framework of distribution testing is currently ubiquitous in the field of property testing. In this model, the input is a probability distribution accessible via independently drawn samples from an oracle. The testing task is to distinguish a distribution that satisfies some property from a distribution that is far in some distance measure from satisfying it. The task of tolerant testing imposes a further restriction, that distributions close to satisfying the property are also accepted. This work focuses on the connection between the sample complexities of non-tolerant testing of distributions and their tolerant testing counterparts. When limiting our scope to label-invariant (symmetric) properties of distributions, we prove that the gap is at most quadratic, ignoring poly-logarithmic factors. Conversely, the property of being the uniform distribution is indeed known to have an almost-quadratic gap. When moving to general, not necessarily label-invariant properties, the situation is more complicated, and we show some partial results. We show that if a property requires the distributions to be non-concentrated, that is, the probability mass of the distribution is sufficiently spread out, then it cannot be non-tolerantly tested with o(root n) many samples, where n denotes the universe size. Clearly, this implies at most a quadratic gap, because a distribution can be learned (and hence tolerantly tested against any property) using O(n) many samples. Being non-concentrated is a strong requirement on properties, as we also prove a close to linear lower bound against their tolerant tests. Apart from the case where the distribution is non-concentrated, we also show if an input distribution is very concentrated, in the sense that it is mostly supported on a subset of size s of the universe, then it can be learned using only O(s) many samples. The learning procedure adapts to the input, and works without knowing s in advance.

关键词： Death penalty   Capital punishment   Education   Politics   Race  

摘要： We argue that the decrease in support for the death penalty associated with college education should differ across political groups given the salience of political identities in the United States. We expect the negative association between education and views of capital punishment is minimized for Republicans and conservatives, especially among White Americans. We use 2002-2018 General Social Survey data and present marginal effects at the mean from the logistic regression models and tests of first and second differences from race-specific interaction models. Moderation analyses demonstrate that the association between education and reduced support for capital punishment is robust among White Democrats and liberals, and relatively weak among White Republicans and especially conservatives. Among non-Whites, the association between higher education and support for capital punishment does not vary by party but it is in the opposite directions for liberals and conservatives. We discuss the implication of the results for party and orientation and for White and non-White respondents, highlighting the rigidity of capital punishment support among political conservatives even as support decreases among other groups.

关键词： Neurons   Visualization   Computational modeling   Artificial neural networks   Backpropagation   Perturbation methods   Heating systems   Decision making   Computer science   Artificial intelligence   Attribution method   deep neural networks   explainable AI   visual explanation  

摘要： With the remarkable success of deep neural networks, there is a growing interest in research aimed at providing clear interpretations of their decision-making processes. In this paper, we introduce Attribution Equilibrium, a novel method to decompose output predictions into fine-grained attributions, balancing positive and negative relevance for clearer visualization of the evidence behind a network decision. We carefully analyze conventional approaches to decision explanation and present a different perspective on the conservation of evidence. We define the evidence as a gap between positive and negative influences among gradient-derived initial contribution maps. Then, we incorporate antagonistic elements and a user-defined criterion for the degree of positive attribution during propagation. Additionally, we consider the role of inactivated neurons in the propagation rule, thereby enhancing the discernment of less relevant elements such as the background. We conduct various assessments in a verified experimental environment with PASCAL VOC 2007, MS COCO 2014, and ImageNet datasets. The results demonstrate that our method outperforms existing attribution methods both qualitatively and quantitatively in identifying the key input features that influence model decisions.

关键词： Black sea bream   Muscle cells   Scaffolds   Three-dimensional culture  

摘要： Culturing fish myogenic cells in vitro holds significant potential to revolutionize aquaculture practices and support sustainable food production. However, advancement in in vitro culture technologies for skeletal muscle-derived myogenic cells have predominantly focused on mammals, with limited studies on fish. Scaffold-based three-dimensional (3D) culture systems for fish myogenic cells remain underexplored, highlighting a critical research gap compared to mammalian systems. This study evaluated the effects of scaffold composition and manufacturing methods on cellular growth in the 3D culture of black sea bream (Acanthopagrus schlegelii) myogenic cells. Scaffolds were manufactured using three natural polymers: black sea bream-derived extracellular matrix (ECM), sodium alginate, and gelatin. Two scaffold types were tested: "cell-laden scaffolds" prepared by mixing cells into the pre-scaffold solution followed by gelation, and "cell-seeding scaffolds" produced by freezing, gelation, and lyophilization before cell inoculation. Scaffold characteristics, including pore size, porosity, swelling ratio, and degradation rate, were assessed. Cell-seeding scaffolds exhibited relatively larger pore size, higher porosity, and higher degradation rate, while cell-laden scaffolds had higher swelling ratios. When black sea bream myogenic cells were cultured in these scaffolds, cell-seeding scaffolds supported cellular growth, particularly when composed of 3% sodium alginate and 4% gelatin with any concentration of ECM. In contrast, cell-laden scaffolds did not support cellular growth regardless of their composition. These findings provide fundamental insights for optimizing scaffold properties to develop more optimized conditions for 3D culture of fish muscle lineage cells.

关键词： automatic question generation   computer science   generative AI   GPT-4   large language models   retrieval-augmented generation  

摘要： BackgroundOver the past few decades, the process and methodology of automatic question generation (AQG) have undergone significant transformations. Recent progress in generative natural language models has opened up new potential in the generation of educational *** paper explores the potential of large language models (LLMs) for generating computer science questions that are sufficiently annotated for automatic learner model updates, are fully situated in the context of a particular course and address the cognitive dimension *** previous attempts that might use basic methods such as ChatGPT, our approach involves more targeted strategies such as retrieval-augmented generation (RAG) to produce contextually relevant and pedagogically meaningful learning *** and ConclusionsOur results show that generating structural, semantic annotations works well. However, this success was not reflected in the case of relational annotations. The quality of the generated questions often did not meet educational standards, highlighting that although LLMs can contribute to the pool of learning materials, their current level of performance requires significant human intervention to refine and validate the generated content.

关键词： Mutual information   Coherence   Training   Semantics   Telecommunications   Computer science   Visualization   Generative models   mutual information (MI)   neural topic modeling   text mining   unsupervised learning  

摘要： To automatically mine structured semantic topics from text, neural topic modeling has arisen and made some progress. However, most existing work focuses on designing a mechanism to enhance topic coherence but sacrificing the diversity of the extracted topics. To address this limitation, we propose the first neural-based topic modeling approach purely based on mutual information maximization, called the mutual information topic (MIT) model, in this article. The proposed MIT significantly improves topic diversity by maximizing the mutual information between word distribution and topic distribution. Meanwhile, MIT also utilizes Dirichlet prior in latent topic space to ensure the quality of mined topics. The experimental results on three publicly benchmark text corpora show that MIT could extract topics with higher coherence values (considering four topic coherence metrics) than competitive approaches and has a significant improvement on topic diversity metric. Besides, our experiments prove that the proposed MIT converges faster and more stable than adversarial-neural topic models.

关键词： TPACK   Scale development   Pre-service teacher   Science education   Programming  

摘要： Programming and computational thinking have been introduced into the curricula of several countries, also in relation to science and technology education. Preparing pre-service teachers for using programming in science education is therefore an important and relevant task. The purpose of this article is to describe what knowledge may be relevant for teachers who are to use programming in science and technology education and to propose a questionnaire to aid in assessing this knowledge. The proposed questionnaire can be used for tracking development over time and for identifying areas where teachers need more knowledge.

关键词： Roads   Accuracy   Data augmentation   Uncertainty   Image segmentation   Deep learning   Real-time systems   Laser radar   Data models   Computer science   Road segmentation   evidence learning   RGB-D data augmentation  

摘要： Despite significant progress in RGB-D based road segmentation in recent years, the latest methods cannot achieve both state-of-the-art accuracy and real time due to the high-performance reliance on heavy structures. We argue that this reliance is due to unsuitable multimodal fusion. To be specific, RGB and depth data in road scenes are each sensitive to different regions, but current RGB-D based road segmentation methods generally combine features within sensitive regions which preserves false road representation from one of the data. Based on such findings, we design an Evidence-based Road Segmentation Method (Evi-RoadSeg), which incorporates prior knowledge of the modal-specific characteristics. Firstly, we abandon the cross-modal fusion operation commonly used in existing multimodal based methods. Instead, we collect the road evidence from RGB and depth inputs separately via two low-latency subnetworks, and fuse the road representation of the two subnetworks by taking both modalities' evidence as a measure of confidence. Secondly, we propose an RGB-D data augmentation scheme tailored to road scenes to enhance the unique properties of RGB and depth data. It facilitates learning by adding more sensitive regions to the samples. Finally, the proposed method is evaluated on the widely used KITTI-road, ORFD, and R2D datasets. Our method achieves state-of-the-art accuracy at over 70 FPS, 5 $\times$ faster than comparable RGB-D methods. Furthermore, extensive experiments illustrate that our method can be deployed on a Jetson Nano 2GB with a speed of 8 $+$ FPS. The code will be released in https://***/xuefeng-cvr/Evi-RoadSeg.