Massive Open Online Courses (MOOCs) have become increasingly popular worldwide. However, learners primarily rely on watching videos, easily losing knowledge context and reducing learning effectiveness. We propose HyperMOOC, a novel approach augmenting MOOC videos with concept-based embedded visualizations to help learners maintain knowledge context. Informed by expert interviews and literature review, HyperMOOC employs multi-glyph designs for different knowledge types and multi-stage interactions for deeper understanding. Using a timeline-based radial visualization, learners can grasp cognitive paths of concepts and navigate courses through hyperlink-based interactions. We evaluated HyperMOOC through a user study with 36 MOOC learners and interviews with two instructors. Results demonstrate that HyperMOOC enhances learners’ learning effect and efficiency on MOOCs, with participants showing higher satisfaction and improved course understanding compared to traditional video-based learning approaches.

Massive Open Online Courses (MOOCs) make high-quality instruction accessible. However, the lack of face-to-face interaction makes it difficult for instructors to obtain feedback on learners' performance and provide more effective instructional guidance. Traditional analytical approaches, such as clickstream logs or quiz scores, capture only coarse-grained learning outcomes and offer limited insight into learners' moment-to-moment cognitive states. In this study, we propose COIVis, an eye-tracking-based visual analytics system that supports concept-level exploration of learning processes in MOOC videos. COIVis first extracts course concepts from multimodal video content and aligns them with the temporal structure and screen space of the lecture, defining Concepts of Interest (COIs), which anchor abstract concepts to specific spatiotemporal regions. Learners' gaze trajectories are transformed into COI sequences, and five interpretable learner-state features-Attention, Cognitive Load, Interest, Preference, and Synchronicity-are computed at the COI level based on eye tracking metrics. Building on these representations, COIVis provides a narrative, multi-view visualization enabling instructors to move from cohort-level overviews to individual learning paths, quickly locate problematic concepts, and compare diverse learning strategies. We evaluate COIVis through two case studies and in-depth user-feedback interviews. The results demonstrate that COIVis effectively provides instructors with valuable insights into the consistency and anomalies of learners' learning patterns, thereby supporting timely and personalized interventions for learners and optimizing instructional design.

In the context of the exponentially increasing volume of narrative texts such as novels and news, readers struggle to extract and consistently remember storylines from these intricate texts due to the constraints of human working memory and attention span. To tackle this issue, we propose a visualization approachStoryExplorer, which facilitates the process of knowledge externalization of narrative texts and further makes the form of mental models more coherent. Through the formative study and close collaboration with two domain experts, we identified key challenges for the extraction of the storyline. Guided by the distilled requirements, we then propose a set of workflow (i.e., insight finding-scripting-storytelling) to enable users to interactively generate fragments of narrative structures. We then propose a visualization systemStoryExplorerthat combines stroke annotation and GPT-based visual hints to quickly extract story fragments and interactively construct storylines. To evaluate the effectiveness and usefulness ofStoryExplorer, we conducted two case studies and in-depth user interviews with 16 target users. The result shows that users can conveniently and effectively extract the storyline by usingStoryExploreralong with the proposed workflow.

Recently, significant progress has been made in condition-guided human motion generation. However, due to the inherent abstraction of conditional semantics like text, music and trajectory, these methods often fall short of generating precise motions that align with human intent. In contrast, free-hand sketches inherently and accurately depict human perspective intent, finding extensive applications across multiple domains. In this article, we introduce Sketch-guided human Motion Diffusion (SMD), to address a novel scenario: sketch-to-motion, aiming to generate plausible and natural human motions based on human motion sketches. Specifically, Our proposed SMD employs a Dual-branch Time-aware Transformer that utilizes both global semantic and local perspective level attention to condition 2D sketch information for 3D motion generation. At the global semantic level, we establish associations between the representation of the entire sketch and the sequential motion to ensure the generated motion aligns with the semantic content of the sketch. Meanwhile, at the local perspective level, a sketch-aware local attention is devised to correlate the sketch patches with the motion keyframes, aiming to precisely align the keyframes with the given sketches. Rooted in Diffusion model and Dual-branch Time-aware Transformer, our approach demonstrates proficiency in motion in-betweening and body part editing tasks, seamlessly generating natural motion sequences that harmonize with the provided context. Multiple experiments conducted on the curated sketch-to-motion datasets validate the efficacy of SMD, showcasing the state-of-the-art generation performances.

Massive Open Online Courses (MOOCs) platforms are becoming increasingly popular in recent years. Online learners need to watch the whole course video on MOOC platforms to learn the underlying new knowledge, which is often tedious and time-consuming due to the lack of a quick overview of the covered knowledge and their structures. In this article, we propose ConceptThread, a visual analytics approach to effectively show the concepts and the relations among them to facilitate effective online learning. Specifically, given that the majority of MOOC videos contain slides, we first leverage video processing and speech analysis techniques, including shot recognition, speech recognition and topic modeling, to extract core knowledge concepts and construct the hierarchical and temporal relations among them. Then, by using a metaphor of thread, we present a novel visualization to intuitively display the concepts based on video sequential flow, and enable learners to perform interactive visual exploration of concepts. We conducted a quantitative study, two case studies, and a user study to extensively evaluate ConceptThread. The results demonstrate the effectiveness and usability of ConceptThread in providing online learners with a quick understanding of the knowledge content of MOOC videos.

Due to its extensive applicability in various downstream domains, few-shot named entity recognition (NER) has attracted increasing attention, particularly in areas where acquiring sufficient labeled data poses a significant challenge. Recent studies have highlighted the potential of knowledge graphs (KGs) in enhancing natural language processing (NLP) tasks. However, a comprehensive understanding of whether and how KGs can effectively improve the NER performance under low-resource conditions remains elusive. In this paper, for the first time, we quantitatively investigate the effects of different kinds of extra KG features for few-shot NER. We enable our analysis by aggregating extra KG features into an NER framework. Through extensive experiments, we find that incorporating class features yields the best performance. To fully explore the potential of class features from KGs, we propose a novel network architecture, named KGen, to jointly leverage KG-based knowledge from both the input sentence side and the label semantic side for few-shot NER.The efficacy of our proposed method is validated through extensive experiments on five challenging datasets.

Parameter-Efficient Fine-Tuning (PEFT) is a promising approach to mitigate the challenges about the model adaptation of pretrained language models (PLMs) for the named entity recognition (NER) task. Recent studies have highlighted the improvements that can be made to the quality of information retrieved from PLMs by adding explicit knowledge from external source like KGs to otherwise naive PEFTs. In this paper, we propose a novel knowledgeable adapter, Know-adapter, to incorporate structure and semantic knowledge of knowledge graphs into PLMs for few-shot NER. First, we construct a related KG entity type sequence for each sentence using a knowledge retriever. However, the type system of a domain-specific NER task is typically independent of that of current KGs and thus exhibits heterogeneity issue inevitably, which makes matching between the original NER and KG types (e.g. Person in NER potentially matches President in KBs) less likely, or introduces unintended noises. Thus, then we design a unified taxonomy based on KG ontology for KG entity types and NER labels. This taxonomy is used to build a learnable shared representation module, which provides shared representations for both KG entity type sequences and NER labels. Based on these shared representations, our Know-adapter introduces high semantic relevance knowledge and structure knowledge from KGs as inductive bias to guide the updating process of the adapter. Additionally, the shared representations guide the learnable representation module to reduce noise in the unsupervised expansion of label words. Extensive experiments on multiple NER datasets show the superiority of Know-Adapter over other state-of-the-art methods in both full-resource and low-resource settings.

Numerous graph clustering methods have been proposed to explore aggregation structures across multiple graphs. In these methods, single-graph features are merely considered or multigraph features are simply weighted, which are insufficient for the construction of reasonable multiple graph clustering features, since the association information between pairwise graphs is ignored and the varied local correlations might influence the clustering preference. Thus, we propose an interactive multiple graph clustering model, iMGC, in this article, to achieve reasonable multiple graph clustering features, which cannot only express multiple relationships, but also preserve associations of nodes across multiple graphs. First, a unified graph matrix is constructed with the combination of structural differences quantified by graph representation learning, which is further optimized by minimizing the difference of structural characteristics between it and each single graph matrix. Thus, multiple relationships are well integrated and expressed, while the varied local correlations within different graphs are also balanced in the unified graph matrix. Then, a constrained Laplacian rank is applied on the unified graph matrix to generate the unified clustering result directly, which is able to preserve association features across multiple graphs. Furthermore, we provide a set of visualization and interaction interfaces, enabling users to intuitively optimize and evaluate the multiple graph clustering features, and interactively explore the multiple graphs. Case studies and quantitative comparisons based on real-world datasets have demonstrated the effectiveness of iMGC in the clustering performance from various perspectives and exploration of multiple graphs.

Numerous sampling strategies have been proposed to reduce the visual clutter of large-scale geographical point data visualization, which focus on the preservation of original data features, such as randomness, spatial distribution, and associated relationship. However, user preferences and demands are not taken into account in the course of sampling, which will lead to the sampled results deviating from user requirements and impede personalized geospatial analysis in specific application scenarios. In this article, we propose a user-driven sampling model for the visual abstraction of the large-scale geographical point data based on convolutional neural networks (CNN). First, a blue noise sampling model is applied to partition the geographical space into local areas, and a set of visual interfaces are designed to present the data features of those points in the local areas, enabling users to visually select representative points according to their requirements. Then, user preferences are quantified with a CNN model based on the eigenvectors of the representative points, which are further utilized to guide the sampling courses of the other local areas. Thus, all the sampled points will retain the spatial distribution of original data points and fulfill the user preference as far as possible. In addition, we implement a visualization framework to integrate manual point selection, CNN training, automatic point sampling, and visual comparison, allowing users to easily obtain and evaluate the sampled points from the perspectives of data analysis and user requirements. Quantitative comparisons and case studies based on real-world datasets are conducted to demonstrate the effectiveness of our sampling model in the preservation of user preferences and visual exploration of large-scale geospatial point data.

Subspace clustering has attracted much attention because of its ability to group unlabeled high-dimensional data into multiple subspaces. Existing graph-based subspace clustering methods focus on either the sparsity of data affinity or the low rank of data affinity. Thus, the quality of data affinity plays an essential role in the performance of subspace clustering. However, the real-world data are generally high-dimensional, complex, and heterogeneous multi-source data, so that the data affinity learned by these methods cannot be completely dependent. Moreover, since these approaches always ignore the intrinsic structure of data, their grouping effect is relatively low. In this paper, we propose a novel unsupervised algorithm, called Structure-Aware Subspace Clustering (SASC), to address the above issues. SASC considers local and global correlation structures simultaneously to capture the intrinsic structure. Further, it integrates the captured structure into representation learning to gain a relatively precise data affinity. It is powerful to promote an all-around grouping effect and enhances the robustness and applicability of subspace clustering. Experiments on various benchmark datasets, including bioinformatics, handwritten digit, object image, and speech signal, demonstrate the effectiveness of the proposed algorithm.

SAR and optical images registration is a key step for remote sensing image processing, match navigation and information fusion. Although there are many methods for SAR images registration, their performance will decrease between SAR and optical images. Moreover, these algorithms suffer from lack of matching pairs of the feature points and uneven distribution between SAR and optical images. Therefore, they cannot accurately achieve the registration between optical and SAR images. To solve the above deficiencies, we propose an efficient image registration approach based on SAR and optical image-scale invariant feature transform (SOI-SIFT). Firstly, a linear edge enhancement based on gray feature and histogram equalization is introduced. In this stage, we enhance the edge features of the image so that the number of image feature points can be greatly increased. Then, for feature points purification, we use fast sample consensus algorithm to filter duplicate and wrong matching feature points. SOI-SIFT can be more adapted to the heterogeneous image matching. Experimental results have shown the superiorities of the proposed method.

For the light microscopy images that have the characteristics of shallow depth of field, serious distortion and poor resolution, mismatch is a ubiquitous phenomenon. The paper presents a mismatch detection method for the stereo light microscopy stereo matching. Affine transformation matrix and matching constraint condition are calibrated by the calibration board which has the precision solid dots and the motorized stage. Bias vector of affine transformation of each matching pair is taken as the criteria to apply mismatch detection. The experimental results show that the method can detect more mismatching pairs and preserve more matching pairs than the traditional RANSAC method and the epipolar rectification method.

In this paper, we propose a novel fast cost propagation algorithm on spanning tree structures. By introducing local smoothness constraint during the weighted cost aggregation process on tree structures, we overcome the shortage of the “fronto-parallel plane” assumption used in most local and non-local cost aggregation algorithms. By applying it to our stereo correspondence framework, accurate results with fine disparity boundaries can be obtained. We also introduce a linear time implementation of the proposed algorithm, thus the computational complexity is kept extremely low.

Abstract To design a bas‐relief from a 3D scene is an inherently interactive task in many scenarios. The user normally needs to get instant feedback to select a proper viewpoint. However, current methods are too slow to facilitate this interaction. This paper proposes a two‐scale bas‐relief modeling method, which is computationally efficient and easy to produce different styles of bas‐reliefs. The input 3D scene is first rendered into two textures, one recording the depth information and the other recording the normal information. The depth map is then compressed to produce a base surface with level‐of‐depth, and the normal map is used to extract local details with two different schemes. One scheme provides certain freedom to design bas‐reliefs with different visual appearances, and the other provides a control over the level of detail. Finally, the local feature details are added into the base surface to produce the final result. Our approach allows for real‐time computation due to its implementation on graphics hardware. Experiments with a wide range of 3D models and scenes show that our approach can effectively generate digital bas‐reliefs in real time.

We propose an angle-based mesh representation, which is invariant under translation, rotation, and uniform scaling, to encode the geometric details of a triangular mesh. Angle-based mesh representation consists of angle quantities defined on the mesh, from which the mesh can be reconstructed uniquely up to translation, rotation, and uniform scaling. The reconstruction process requires solving three sparse linear systems: the first system encodes the length of edges between vertices on the mesh, the second system encodes the relationship of local frames between two adjacent vertices on the mesh, and the third system defines the position of the vertices via the edge length and the local frames. From this angle-based mesh representation, we propose a quasi-angle-preserving mesh deformation system with the least-squares approach via handle translation, rotation, and uniform scaling. Several detail-preserving mesh editing examples are presented to demonstrate the effectiveness of the proposed method.

Abstract This paper presents a novel modeling system, called B‐Mesh, for generating base meshes of 3D articulated shapes. The user only needs to draw a one‐dimensional skeleton and to specify key balls at the skeletal nodes. The system then automatically generates a quad dominant initial mesh. Further subdivision and evolution are performed to refine the initial mesh and generate a quad mesh which has good edge flow along the skeleton directions. The user can also modify and manipulate the shape by editing the skeleton and the key balls and can easily compose new shapes by cutting and pasting existing models in our system. The mesh models generated in our system greatly benefit the sculpting operators for sculpting modeling and skeleton‐based animation.

The stretching energy functional and the bending energy functional are widely used for approximating the solution of the Plateau-Béizer Problem. This paper presents another two simple methods by using the extended stretching energy functional and the extended bending energy functional. The resulting surface obtained by the new methods will have a smaller area. Comparisons are made with both the area and the mean curvature of the resulting surfaces.

In this paper, three iterative learning control (ILC) schemes are developed in the multirate signal processing domain. One is pseudo-downsampled ILC, in which the input update rate is different from the sampling rate of feedback system. The second one is a two-mode ILC, in which the input update rates of ILC are different at low and high frequency bands. The third one is a cyclic pseudo-downsampled ILC, which extends the first scheme by shifting downsampling points in different iterations. Theoretical background and design approaches of these multirate schemes are addressed. Experimental results are presented to highlight the traits of each scheme. The advantage is that these schemes have the ability to learn those error component beyond the learnable bandwidth of a conventional ILC and, therefore, can improve the tracking accuracy substantially. Additionally, the multirate ILC schemes have the abilities to produce good learning transient with the presence of initial state error.

The size-driven mechanism is only applied in automatically reconstructing parts by most of recent popular 3D CAD tools, however, most of them have not realized the automatic reconstructing assemblies at all. In this paper, we propose a novel 3D modeling method, which integrates size parameter and change rules into the model, so that an assembly becomes an object which can automatically reconstruct any parts according to the rules once the whole model size changes. Using this method, we not only keep the size- driven mechanism in part, but also make assemblies have the ability of automatic reconstruction. The results of experiments in furniture design application show our method can greatly improve models' flexibility and reusability.

The earlier works by Zhang, B. et al, (2004) on cutoff-frequency phase-in ILC show that the scheme can suppress initial state error/position offset properly and improve tracking accuracy. However, since cutoff frequency is set high in initial phase of operation cycles, the improvement of tracking accuracy is mainly in this phase and the tracking error in later phase of operation cycles can still be large. A uniformly good tracking accuracy is favorable in many applications. In this paper, a sliding cutoff-frequency phase-in ILC is proposed to achieve this goal. In this scheme, cutoff frequency profile moves along the time axis after some cycles according to the assessment of tracking performance. Experimental results on an SCARA robot show that this scheme can further improve the tracking accuracy and generate a uniform tracking error over the entire operation interval

This paper proposes a new preprocessing method for interactive rendering of complex polygonal virtual environments. The approach divides the space that observer can reach into many rectangular viewpoint regions. For each region, an outer rectangular volume (ORV) is established to surround it. By adaptively partitioning the boundary of the ORV together with the viewpoint region, all the rays that originate from the viewpoint region are divided into the beams whose potentially visible polygon number is less than a preset threshold. If a resultant beam is the smallest and intersects many potentially visible polygons, the beam is simplified as a fixed number of rays and the averaged color of the hit polygons is recorded. For other beams, their potentially visible sets (PVS) of polygons are stored respectively. During an interactive walkthrough, the visual information related to the current viewpoint is retrieved from the storage. The view volume clipping, visibility culling and detail simplification are efficiently supported by these stored data. The rendering time is independent of the scene complexity.