With the rapid development of computer vision, 3D data is increasing rapidly. How to retrieve similar model from a large number of models has become a hot research topic. However, in order to meet people's demand, the retrieval accuracy need to be further improved. In terms of multi-view 3D model retrieval, how to effectively learn the information between views is the key to improving performance. In this paper, we propose a novel 3D model retrieval algorithm based on attention and multi-view fusion. Specifically, we mainly constructed two modules. First, dynamic attentive graph learning module is used to learn the intrinsic relationship between view blocks; Then we propose the Attention-NetVlad algorithm, which combines the channel attention algorithm and the NetVlad algorithm. It learns the information between feature channels to enhance the feature expression ability firstly, then uses the NetVlad algorithm to fuse multiple view features into a global feature according to the clustering information. Finally the global feature is used as the only feature of the model to retrieve according to Euclidean distance. In comparison with other state-of-the-art methods by utilizing ModelNet10 and ModelNet40 the proposed method has demonstrated significant improvement for retrieval mAP. Our experiments also demonstrate the effectiveness of the modules in the algorithm.
Accurate detection of extremely small targets in aerial high-resolution images is critical for military and civilian applications. However, challenges such as small target size, complex backgrounds, and target deformation hinder optimal performance. We propose QLDNet, a lightweight network that addresses these issues through a synergistic combination of non-strided convolution and decoupled large-kernel convolutional attention. Specifically, the non-strided convolution constructs a quadruple aggregate and connection feature extractor (QACFE) module, which maps features into the channel dimension while preserving spatial details. The decoupled large kernel convolutional attention then leverages these channel features to effectively extract structural and edge-related low-frequency information, while simultaneously reducing computational costs and model size caused by the increased channel dimensionality after QACFE. A learnable offset mechanism is introduced, transforming the detection head into a deformable detection head. Additionally, the network incorporates a PAFPN (Path Aggregation Feature Pyramid Network) structure to efficiently extract multi-scale features. During inference, a tailored approach performs pixel-level multi-scale detection. This enhances small target detection by merging features and pixels through dual-scale fusion, integrating multi-scale feature extraction with multilevel feature integration.Experimental results demonstrate that QLDNet achieves efficient and accurate small target detection, with an accuracy rate as high as 94.0\% and an inference time of 0.12 seconds per image, meeting the system's real-time requirements with fewer parameters and satisfying processing speeds.The code can be find at https://github.com/Sjl185721/QLDNet-v1.git. Mathematics Subject Classification (2020) MSC68T05 · MSC 86Axx · MSC 68T10
29 August 2025 in Journal of Real-Time Image Processing