Offsetting-based hollowing is a solid modeling operation widely used in 3D printing, which can change the model's physical properties and reduce the weight by generating voids inside a model. However, a hollowing operation can lead to additional supporting structures for fabrication in interior voids, which cannot be removed. As a consequence, the result of a hollowing operation is affected by these additional supporting structures when applying the operation to optimize physical properties of different models. This paper proposes a support-free hollowing framework to overcome the difficulty of fabricating voids inside a solid. The challenge of computing a support-free hollowing is decomposed into a sequence of shape optimization steps, which are repeatedly applied to interior mesh surfaces. The optimization of physical properties in different applications can be easily integrated into our framework. Comparing to prior approaches that can generate support-free inner structures, our hollowing operation can reduce more volume of material and thus provide a larger solution space for physical optimization. Experimental tests are taken on a number of 3D models to demonstrate the effectiveness of this framework.

Designing an effective and uniform framework to meliorate tracking performance is very meaningful and essential. To tackle this issue, in this paper, we present an informative enhanced loss based Siamese tracking network. Specifically, as existing methods merely focus on the single positive and negative instances corresponding to the exemplar, they hence thoroughly ignore the effective information hidden in other instances. To this end, we propose an informative enhanced loss to enable the network to capture information from an overall perspective. In other words, we construct dense connections among instances and exemplar to take full advantage of information from these samples. More importantly, we prove that our proposed loss can be transformed into the logistic loss and the triplet loss under particular parameter settings. Experiments on prevalent benchmarks demonstrate that the Siamese frameworks trained with our proposed loss indeed obtain better tracking results than original methods and achieve promising performance against some state-of-the-art trackers on the real-time challenge.

Entrepreneurship education is critical in encouraging students' innovation, creativity, and entrepreneurial spirit. It provides essential skills and knowledge, enabling them to open their creative potential and apply innovative thinking across diverse professional fields. With the widespread application of large language models in education, intelligent-assisted teaching in entrepreneurship education is stepping into a new learning phase anytime and anywhere. Entrepreneurship education extends across interdisciplinary knowledge fields, incorporating subjects like finance and risk management, which require advanced mathematical computational skills. This complexity presents new challenges for artificial intelligence-assisted question-and-answer models. The study explores how students can maximize the knowledge repository of current large language models to improve learning efficiency. Experimentally validates the performance differences between large language models and graph convolutional reasoning models regarding the complex semantic reasoning and mathematical computational demands in entrepreneurship education questions. Based on case studies, it is found that despite the broad prospects of large language models in entrepreneurship education, they still need to improve in practical applications. Especially in tasks within entrepreneurship education that demand precision, such as mathematical computations and risk assessment, the accuracy and efficiency of existing models still need improvement. Therefore, further exploration into algorithm optimization, model fusion, and other technical enhancements can improve the processing capabilities of intelligent question-and-answer systems for specific domain issues, aiming to meet the practical needs of entrepreneurship education.

The widespread deployment of Deep Neural Networks (DNNs) for 3D point cloud processing contrasts sharply with their vulnerability to security breaches, particularly backdoor attacks. Studying these attacks is crucial for raising security awareness and mitigating potential risks. However, the irregularity of 3D data and the heterogeneity of 3D DNNs pose unique challenges. Existing methods frequently fail against basic point cloud preprocessing or require intricate manual design. Exploring simple, imperceptible, effective, and difficult-to-defend triggers in 3D point clouds remains challenging. To address this issue, we propose iBA, a novel solution utilizing a folding-based auto-encoder (AE). By leveraging united reconstruction losses, iBA enhances both effectiveness and imperceptibility. Its data-driven nature eliminates the need for complex manual design, while the AE core imparts significant nonlinearity and sample specificity to the trigger, rendering traditional preprocessing techniques ineffective. Additionally, a trigger smoothing module based on spherical harmonic transformation (SHT) allows for controllable intensity. We also discuss potential countermeasures and the possibility of physical deployment for iBA as an extensive reference. Both quantitative and qualitative results demonstrate the effectiveness of our method, achieving state-of-the-art attack success rates (ASR) across a variety of victim models, even with defensive measures in place. iBA’s imperceptibility is validated with multiple metrics as well.

Although recent Siamese network-based trackers have achieved impressive perceptual accuracy for single object tracking in LiDAR point clouds, they usually utilized heavy correlation operations to capture category-level characteristics only, and overlook the inherent merit of arbitrariness in contrast to multiple object tracking. In this work, we propose a radically novel one-stream network with the strength of the instance-level encoding, which avoids the correlation operations occurring in previous Siamese network, thus considerably reducing the computational effort. In particular, the proposed method mainly consists of a Template-aware Transformer Module (TTM) and a Multi-scale Feature Aggregation (MFA) module capable of fusing spatial and semantic information. The TTM stitches the specified template and the search region together and leverages an attention mechanism to establish the information flow, breaking the previous pattern of independent extraction-and-correlation . As a result, this module makes it possible to directly generate template-aware features that are suitable for the arbitrary and continuously changing nature of the target, enabling the model to deal with unseen categories. In addition, the MFA is proposed to make spatial and semantic information complementary to each other, which is characterized by reverse directional feature propagation that aggregates information from shallow to deep layers. Extensive experiments on KITTI and nuScenes demonstrate that our method has achieved considerable performance not only for class-specific tracking but also for class-agnostic tracking with less computation and higher efficiency.