Multimodal LiDAR Semantic Segmentation Based on Adaptive Collaborative Learning

Adaptation and personalization have been extensively studied in CSCL research community aiming to design intelligent systems that adaptively support eLearning processes and collaboration. Yet, with the fast development in Internet technologies, especially with the emergence of new data technologies and the mobile technologies, new opportunities and perspectives are opened for advanced adaptive and personalized systems. Adaptation and personalization are posing new research and development challenges to nowadays CSCL systems. In particular, adaptation should be focused in a multi-dimensional way (cognitive, technological, context-aware and personal). Moreover, it should address the particularities of both individual learners and group collaboration. As a consequence, the aim of this book is twofold. On the one hand, it discusses the latest advances and findings in the area of intelligent adaptive and personalized learning systems. On the other hand it analyzes the new implementation perspectives for intelligent adaptive learning and collaborative systems that are brought by the advances in scripting languages, IMS LD, educational modeling languages and learning activity management systems. Given the variety of learning needs as well as the existence of different technological solutions, the book exemplifies the methodologies and best practices through several case studies and adaptive real-world collaborative learning scenarios, which show the advancement in the field of analysis, design and implementation of intelligent adaptive and personalized systems.

Collaborative learning in virtual classrooms has tremendous potential in institutions, as it encourages cooperative knowledge production and skill development. This study investigates how a pedagogical model for virtual learning can incorporate this methodology, emphasizing the significance of meticulous preparation, an appropriate dynamic for establishing groups, the relevance of student practices to everyday technology utilization, an evolution to educator responsibilities and learning autonomy. The research underlines the value of technology in education. This article discusses Virtual Collaborative Learning (VCL) as an effective approach to quality assurance in the age of digitization. It presents design characteristics from an academic standpoint and optimizes these approaches through a qualitative examination of written opinions by VCL participants. The research identifies and prioritizes critical criteria for collaborative learning effectiveness from the students' perspective, generating further multi-perspective design recommendations. Adaptive collaborative virtual learning is a technology-enabled technique that employs algorithms to evaluate student data and adjust to a student's learning style, pace, and accomplishments. It capitalizes on gamification to make learning more engaging and interactive, while artificial intelligence (AI), machine learning, virtual reality (VR), and augmented reality (AR) have transformed conventional education. This article explores at how cutting-edge technologies are being created and implemented into the educational system and classroom for enhancing student education and learning environments. This chapter examines the adoption of AI and machine learning in intelligent learning, emphasizing their potential for enhancing learning experiences, personalize education, and improve outcomes. It also encompasses ethical and security issues, highlighting the importance of stringent laws to protect students’ rights. By implementing these guidelines, educators and policymakers may build a more intelligent and successful classroom setting. This study visualizes AI-enabled learning systems, identifies types of interventions, and discusses popular analytical methodologies. It serves as a reference for future studies regarding the development of AI-enabled educational platforms that address specific learning challenges while enhancing user experiences, ultimately directing future research in this area of study.

Multimodal image fusion and semantic segmentation are critical for autonomous driving. Despite advancements, current models often struggle with segmenting densely packed elements due to a lack of comprehensive fusion features for guidance during training. While the Segment Anything Model (SAM) allows precise control during fine-tuning through its flexible prompting encoder, its potential remains largely unexplored in the context of multimodal segmentation for natural images. In this paper, we introduce SAM into multimodal image segmentation for the first time, proposing a novel framework that combines Latent Space Token Generation (LSTG) and Fusion Mask Prompting (FMP) modules. This approach transforms the training methodology for multimodal segmentation from a traditional black-box approach to a controllable, prompt-based mechanism. Specifically, we obtain latent space features for both modalities through vector quantization and embed them into a cross-attention-based inter-domain fusion module to establish long-range dependencies between modalities. We then use these comprehensive fusion features as prompts to guide precise pixel-level segmentation. Extensive experiments on multiple public datasets demonstrate that our method significantly outperforms SAM and SAM2 in multimodal autonomous driving scenarios, achieving an average improvement of 4.1% over the state-of-the-art method in segmentation mIoU, and the performance is also optimized in other multi-modal visual scenes.

A key promise of adaptive collaborative learning support is the ability to improve learning outcomes by providing individual students with the help they need to collaborate more effectively. These systems have focused on a single platform. However, recent technology-supported collaborative learning platforms allow students to collaborate in different contexts: computer-supported classroom environments, network based online learning environments, or virtual learning environments with pedagogical agents. Our goal is to better understand how students participate in collaborative behaviors across platforms, focusing on a specific type of collaboration - help-giving. We conducted a classroom study (N = 20) to understand how students engage in help-giving across two platforms: an interactive digital learning environment and an online Q&A community. The results indicate that help-giving behavior across the two platforms is mostly influenced by the context rather than by individual differences. We discuss the implications of the results and suggest design recommendations for developing an adaptive collaborative learning support system that promotes learning and transfer.

The paper proposes a methodology that emphasis techno-pedagogy, namely the constructivist and the adaptive intelligent learning of specialized semiology of COVID-19. An e-learning built on a constructivist pedagogy with a technology such adaptive intelligent environment, can be individualized (adaptive learning), can enhance learners' interactions with others (collaborative learning), and transforming the role of the teacher as facilitators of learning and assessor of competency. To make our system intelligent, we cope with Artificial Intelligence and Big data.

Personalized context-aware systems for sustainable agriculture development using ubiquitous devices and adaptive learning

Point cloud segmentation is essential for scene understanding, which provides advanced information for many applications, such as autonomous driving, robots, and virtual reality. To improve the accuracy and robustness of point cloud segmentation, many researchers have attempted to fuze camera images to complement the color and texture information. The common fusion strategy is the combination of convolutional operations with concatenation, element-wise addition or element-wise multiplication. However, conventional convolutional operators tend to confine the fusion of modal features within their receptive fields, which can be incomplete and limited. In addition, the inability of encoder–decoder segmentation networks to explicitly perceive segmentation boundary information results in semantic ambiguity and classification errors at object edges. These errors are further amplified in point cloud segmentation tasks, significantly affecting the accuracy of point cloud segmentation. To address the above issues, we propose a novel self-attention multi-modal fusion semantic segmentation network for point cloud semantic segmentation. Firstly, to effectively fuze different modal features, we propose a self-cross fusion module (SCF), which models long-range modality dependencies and transfers complementary image information to the point cloud to fully leverage the modality-specific advantages. Secondly, we design the salience refinement module (SR), which calculates the importance of channels in the feature maps and global descriptors to enhance the representation capability of salient modal features. Finally, we propose the local-aware anisotropy loss measure the element-level importance in the data and explicitly provide boundary information for the model, which alleviates the inherent semantic ambiguity problem in segmentation networks. Extensive experiments on two benchmark datasets demonstrate that our proposed method surpasses current state-of-the-art methods.

The application of multimodal image fusion has become increasingly widespread across various fields in the era of deep learning. Existing fusion methods integrate infrared and visible images to provide complementary content and enhance the robustness of complex real-world scenes for high-level visual tasks, such as semantic segmentation and object detection. In return, high-level visual tasks facilitate the fusion of infrared and visible by providing mid-level semantic information. However, such frameworks rely heavily on multimodal data and require strict registration of images from different modalities before fusion, seriously limiting their practical applications due to the common realistic situations of missing modalities or misregistration. To move beyond this limitation, we propose a novel hierarchical knowledge distillation (HKD) framework tailored for unimodal image segmentation with the guidance of multi-modality. This framework aims to retain as much diverse information from multimodal image fusion as possible, thereby enhancing downstream high-level visual tasks when only the unimodal images are available during the inference phase. Our proposed method is two-stage, and we construct a robust multimodal fusion and segmentation interaction network in the first stage as a powerful teacher model. In the second stage, we design a hierarchical distillation method to transfer the fused and segmented multi-layer knowledge from the multimodal teacher model to the unimodal student model. Extensive experimental results on two public datasets, i.e., MFNet and FMB, demonstrate that the proposed hierarchical knowledge distillation framework can effectively transfuse multimodal knowledge into the unimodal student model for image enhancement and segmentation under incomplete multimodal conditions, and achieves considerably competitive results compared to multimodal image fusion and segmentation models.

The rapid advancement with regard to Artificial Intelligence (AI) has transformed multiple educational domains, particularly in adaptive teaching and learning, where AI technologies play a crucial role in personalizing instruction and enhancing learning outcomes. Despite its increasing significance, there remains a lack of comprehensive bibliometric evidence that maps the intellectual structure, research patterns, and emerging trends in this field. This study addresses this gap by conducting a bibliometric analysis of AI-assisted adaptive teaching and learning to identify publication dynamics, research hotspots, and global collaboration networks. Data were collected from the Scopus database via advanced search strategies incorporating the keywords “Artificial Intelligence,” “Adaptive,” “Teaching,” and “Learning,” which yielded 866 relevant publications. The dataset was cleaned and harmonized using OpenRefine to ensure consistency, while statistical and graphical analyses were performed through the Scopus analyzer. Visualization of co-authorship, keyword co-occurrence, and thematic clustering was generated using VOSviewer to offer a deeper comprehension regarding the field’s conceptual and intellectual structures. The results reveal a significant growth trajectory of publications over the past decade, dominated by contributions from China, the United States, India, and the United Kingdom, highlighting the global nature of research in this area. Keyword co-occurrence analysis identified core themes, for example, intelligent tutoring systems, machine learning (ML), adaptive learning environments, and collaborative learning, with emerging interest in areas like deep learning, educational data mining, and personalized learning pathways. These research findings demonstrate that AI-assisted adaptive teaching and learning research is highly interdisciplinary, bridging education, computer science, and cognitive sciences, and is increasingly shaped by international collaborations. In conclusion, this study provides an overview of the knowledge structure in AI-driven adaptive education and offers valuable insights for scholars, practitioners, and policymakers. These insights help identify future research directions, strengthen cross-border partnerships, and advance innovative pedagogical frameworks that leverage AI for inclusive and effective learning.

Multi-modal 3D semantic segmentation is vital for applications such as autonomous driving and virtual reality (VR). To effectively deploy these models in real-world scenarios, it is essential to employ cross-domain adaptation techniques that bridge the gap between training data and real-world data. Recently, self-training with pseudo-labels has emerged as a predominant method for cross-domain adaptation in multi-modal 3D semantic segmentation. However, generating reliable pseudo-labels necessitates stringent constraints, which often result in sparse pseudo-labels after pruning. This sparsity can potentially hinder performance improvement during the adaptation process. We propose an image-guided pseudo-label enhancement approach that leverages the complementary 2D prior knowledge from the Segment Anything Model (SAM) to introduce more reliable pseudo-labels, thereby boosting domain adaptation performance. Specifically, given a 3D point cloud and the SAM masks from its paired image data, we collect all 3D points covered by each SAM mask that potentially belong to the same object. Then our method refines the pseudo-labels within each SAM mask in two steps. First, we determine the class label for each mask using majority voting and employ various constraints to filter out unreliable mask labels. Next, we introduce Geometry-Aware Progressive Propagation (GAPP) which propagates the mask label to all 3D points within the SAM mask while avoiding outliers caused by 2D-3D misalignment. Experiments conducted across multiple datasets and domain adaptation scenarios demonstrate that our proposed method significantly increases the quantity of high-quality pseudo-labels and enhances the adaptation performance over baseline methods.

Accurate semantic segmentation of high-resolution remote sensing imagery is challenged by substantial intra-class variability, inter-class similarity, and the limitations of single-modality data. This paper proposes MMFNet, a novel multimodal fusion network that leverages the Mamba architecture to efficiently capture long-range dependencies for semantic segmentation tasks. MMFNet adopts a dual-encoder design, combining ResNet-18 for local detail extraction and VMamba for global contextual modelling, striking a balance between segmentation accuracy and computational efficiency. A Multimodal Feature Fusion Block (MFFB) is introduced to effectively integrate complementary information from optical imagery and digital surface models (DSMs), thereby enhancing multimodal feature interaction and improving segmentation accuracy. Furthermore, a frequency-aware upsampling module (FreqFusion) is incorporated in the decoder to enhance boundary delineation and recover fine spatial details. Extensive experiments on the ISPRS Vaihingen and Potsdam benchmarks demonstrate that MMFNet achieves mean IoU scores of 83.50% and 86.06%, outperforming eight state-of-the-art methods while maintaining relatively low computational complexity. These results highlight MMFNet's potential for efficient and accurate multimodal semantic segmentation in remote sensing applications.

Human Computer Interaction Semantic filtering techniques are used in learning environment to track problems in collaborative systems. However, as noted in Adigun et al. [1], when sharing and dynamism are promoted, a problem of redundancy and integrity appeared not to have been well addressed. An improved ASF-based method of evaluating semantic filtering for social network systems in collaborative learning environment is developed, which assisted participants to achieve greater levels of performance with information sharing from other collaborators, as well as in reusing ideas across the period of collaboration.

Adaptive Collaborative Learning Support (ACLS) systems improve collaboration and learning for students over individual work or collaboration with non-adaptive support. However, many ACLS systems are ill-suited for rural contexts where students often need multiple kinds of support to complete tasks, may speak languages unsupported by the system, and require more than pre-assigned tutor-tutee student pairs for more equitable learning. We designed an intervention that fosters more equitable help-seeking by automatically detecting student struggles and prompts them to seek help from specific peers that can help. We conducted a mixed-methods experimental study with 98 K-3 students in a rural village in Tanzania over a one-month period, evaluating how the system affects student interactions, system engagement, and student learning. Our intervention increased student interactions by almost 4 times compared to the control condition, increased domain knowledge interactions, and propelled students to engage in more cognitively challenging activities.

Adaptive Collaborative Learning Process in a Hybrid Model

University Level Learning and Teaching via E-Learning Platforms