Infrared Small Target Detection (IRSTD) requires strong target-level detection capability, which depends on effective modeling of long-range global dependencies. This demand has driven the transition from CNN-based approaches to Transformer-based architectures. Although Transformers improve global context modeling, their high computational cost limits practical deployment. Recent advances in Mamba enable efficient long-range dependency modeling with reduced complexity, offering a promising alternative that alleviates the efficiency limitations of Transformers while preserving target-level detection performance. However, Mamba is not inherently tailored for IRSTD, as it lacks explicit mechanisms for capturing fine-grained local details and modeling background variations across multiple spatial scales. To address these limitations, we propose MCFNet, an encoder-decoder framework that integrates Mamba to enhance target-level detection performance with moderate computational cost. MCFNet introduces a Detail-Capturable Convolution Block to strengthen local detail perception and a Multi-scale Contextual Mamba Block to improve background modeling across different scales. While the resulting dual-branch design enhances both global semantics and local details, it also introduces challenges in feature fusion. To this end, a Feature Fusion Decoding Module is further proposed to enable effective collaboration between global and local representations. Extensive experiments on multiple public IRSTD benchmark datasets demonstrate that MCFNet consistently outperforms existing methods in both pixel-level and target-level metrics, achieving higher detection accuracy with reduced false alarms. The code of our model is available at: https://github.com/Fihven/MCFNet.

Fine-grained 3D shape classification (FGSC) poses a unique challenge due to subtle inter-class differences and intricate structural information, often leading to suboptimal performance. This paper introduces GGView-Net, a novel approach for FGSC utilizing a progressive Glance and Gaze mechanism inspired by human object recognition processes. GGView-Net comprises three key stages - Glance, Gaze, and Joint Classification. The Glance phase mimics the holistic perception of a 3D shape by extracting global features, while the Gaze stage focuses on generating intra-view attention features guided by the global context. Finally, the Joint Classification step tailors losses to individual views and FGSC requirements. Through comprehensive exper-iments on FG3D, GGView-Net demonstrates superior FGSC performance using only class labels, with visualization results confirming its interpretability. Moreover, GGView-Net achieves state-of-the-art results in meta-category 3D shape classification on the ModelNet40 dataset, showcasing remarkable versatility.

Large-scale pre-trained models have demonstrated significant advantages when dealing with visual and language tasks in open-world scenes. However, recent studies have identified certain limitations when utilizing comparative language-visual pretrained models for zero-shot 3D model classification, mainly in the form of neglecting the multi-view independent information of the 3D model. In addition, existing textual cues usually rely only on category semantics and fail to fully exploit the rich contextual structure of the 3D model itself, thus limiting the effective matching of visual and semantic information. To this end, this paper proposes a multi-view text-enhanced zero-shot 3D model classification based on a parameter-free network. The method utilizes a pre-trained image encoder to extract multi-view visual features and combines the semantic comprehension capability of a large language model to mine 3D model structure information from multiple perspectives. Fine-grained associations between views and textual cues are made through view-by-view interaction, and decision-level fusion is used to integrate the independent decision results of each view to achieve zero-shot classification. Without any 3D training, the method in this paper achieves 67.4% accuracy on the ZS3D dataset and 61.3%, 43.3% and 28.1% accuracy on the three sub-datasets of the Ali dataset, which verifies the generalizability and effectiveness of the proposed method.

Fine-grained 3D shape classification (FGSC) remains challenging due to the difficulty of adaptively capturing global structure differences and subtle inter-class distinctions. This paper directly extends Vision Transformer (ViT) to FGSC, proposing a pure Transformer network FG3DFormer that fully leverages ViT’s global correlation and local attention abilities. FG3Dformer comprises the Hierarchical Feature Extraction (HFE) and the Hierarchical Feature Refinement (HFR), interconnected through the Adaptive View Region Selection (AVRS). Firstly, the HFE comprehensively evaluates the significance of intra-view patches and views driven by inter-view and intraview attention. Then, the AVRS adaptively selects crucial patch Tokens from different views to serve as sources of subtle local features. Finally, the HFR refines the 3D shape descriptor, capturing more discriminative global and subtle local features by leveraging both the view and selected crucial patch Tokens. Extensive experiments on FG3D and ModelNet40 demonstrate the superiority of FG3Dformer in FGSC and meta-category 3D shape classification tasks.

Sketch-based 3D shape retrieval has become a prominent area of research in computer vision, confronting challenges related to the inherent diversity and abstraction of sketches, as well as inter-domain discrepancies. This paper introduces a novel approach called Dual Knowledge Distillation Dynamic Learning (DKD2L) aimed at enhancing the extraction of spatio-temporal features for sketch-based 3D shape retrieval. We develop a temporal feature extraction network to effectively capture the dynamic temporal characteristics of sketches and improve retrieval efficiency through temporal knowledge distillation. Additionally, to tackle intra-class variation and inter-class imbalance, we apply semantic knowledge distillation, enabling the 3D shape network to guide the sketch network in capturing common semantic information. This approach facilitates precise cross-modal alignment and enhances retrieval accuracy. Extensive experiments on two benchmark datasets demonstrate that DKD2L surpasses existing state-of-the-art methods.

Sketch-based 3D shape retrieval (SBSR) is an active research area in the computer vision community, but it is still very challenging. One main reason is that existing methods usually treat sketches as regular images, neglecting their unique abstraction and sparsity. This paper introduces a novel SBSR Transformer framework, D2Former, which fully exploits the uniqueness of hand-drawn sketches composed of strokes. For sketches, we first decompose the dynamical sketch into a sequence of partially overlapping sub-sketches and design a Dynamic Sub-sketch Select Module (DSSM) to select key sub-sketch features. The Transformer architecture is then employed to integrate global sketch features while highlighting their representative discriminative features. For 3D shapes, we construct an Inter-View Fusion Module (IVFM), which can incorporate features from multiple views and accurately capture the associations between them. Experiments on two benchmark datasets demonstrate that our method exhibits superior retrieval performance compared to the current state-of-the-art SBSR methods.

Fine-grained 3D shape classification (FGSC) has garnered significant attention recently and has made notable advancements. However, due to high inter-class similarity and intra-class diversity, it is still a challenge for existing methods to capture subtle differences between different subcategories for FGSC. On the one hand, one-hot labels in loss function are too hard to describe the above data characteristics, and on the other hand, local details are submerged in the global features extraction process and final network constraints, impacting classification results. In this paper, we propose a duplex label smoothing-based hierarchical context-aware network for fine-grained 3D shape classification, named DLS-HCAN. Specifically, DLS-HCAN firstly employs a hierarchical context-aware network (HCAN), in which the intra-view context attention mechanism (intra-ATT) and the inter-view context multilayer perceptron (inter-MLP) are designed to focus on and discern the beneficial local details. Subsequently, we propose a novel duplex label smoothing (DLS) regularization in which shape-level and view-level smooth labels are separately applied in two improved loss functions, adapting to the fine-grained data characteristics and considering the varying uniqueness of different views. Notably, our approach does not require additional annotation information. Experimental results and comparison with state-of-the-art methods demonstrate the superiority of our proposed DLS-HCAN for FGSC. In addition, our approach also achieves comparable performance for the coarse-grained dataset on ModelNet40.

Many rapid fabrication technologies are directed towards layer wise printing or laser based prototyping. We propose WireFab, a rapid modeling and prototyping system that uses bent metal wires as the structure framework. WireFab approximates both the skeletal articulation and the skin appearance of the corresponding virtual skin meshes, and it allows users to personalize the designs by (1) specifying joint positions and part segmentations, (2) defining joint types and motion ranges to build a wire-based skeletal model, and (3) abstracting the segmented meshes into mixed-dimensional appearance patterns or attachments.

In this paper, we present a semantic-based partial retrieval approach of CAD models for design reuse. Based on the observation of reusable regions for design of 3D CAD models, for each model in the model library, we propose an algorithm that automatically extracts its reusable regions for partial retrieval. To further effectively support the partial retrieval of these reusable regions through simple queries, we represent each reusable region by all its local matching regions and describe each local matching region in a hierarchical way. Based on the hierarchical descriptor, a partial retrieval method for reusable regions is put forward. The approach proposed is implemented and tested by hundreds of mechanical parts. Preliminary results show that the method can effectively support partial retrieval for design reuse.

In this paper, a multi-mode retrieval approach of solid models is presented. First, a local feature determination method based on wavelet transform is given, which makes the local feature agree with the human perception of resemblance; then a uniform and efficient DBMS (dilation based multiresolutional skeleton) generation method for global model and local feature using adaptive voxelization and local dilation is proposed; finally a flexible similarity assessment is proposed which supports both global matching and partial matching effectively and further supports different retrieval requires. Preliminary experiment results show that a multi-mode retrieval approach can describe user's retrieval intent more precisely and return retrieval results more properly.