常到 发表于 2025-3-27 00:16:10
http://reply.papertrans.cn/47/4615/461478/461478_31.pngDorsal 发表于 2025-3-27 01:26:24
http://reply.papertrans.cn/47/4615/461478/461478_32.pngreptile 发表于 2025-3-27 09:19:24
MGP-Net: Margin-Global Information Optimization-Prototype Network for Few-Shot Ancient Inscriptions Traditional classification methods have failed to produce satisfactory results. With the emergence of meta-learning, few-shot image classification has become a popular research topic. This approach allows a classifier to recognize datasets outside the training set and complete classification with on纵火 发表于 2025-3-27 12:03:40
http://reply.papertrans.cn/47/4615/461478/461478_34.png狗窝 发表于 2025-3-27 16:52:37
Foreign Object Detection Based on Compositional Scene Modelinging, and color difference in scenes in industrial production, it is difficult for many anomaly detection methods to maintain high performance. We propose a foreign object detection method based on compositional scene modeling, which includes a background denoising module and a foreground reconstructJogging 发表于 2025-3-27 20:12:29
http://reply.papertrans.cn/47/4615/461478/461478_36.pngBravura 发表于 2025-3-28 00:10:49
Disentangled Shape and Pose Based on Attention and Mesh Autoencoderr approach builds a new mesh auto-encoder based on the AFA module for effective feature aggregation. We demonstrate that our proposed A-MeshNet achieves superior performance than state-of-the-art approaches on various benchmarking datasets.Grievance 发表于 2025-3-28 05:13:46
http://reply.papertrans.cn/47/4615/461478/461478_38.pngZEST 发表于 2025-3-28 06:18:39
Dense Small Object Detection Based on Improved Deep Separable Convolution YOLOv5 model is improved. Attention mechanism and lightweight module supplement each other, enables the attention mechanism to work better. Simulation experiments were conducted on CrowdHuman dataset, and the experimental results showed that YOLOv5-G increased by 3.1% compared with the original YOLOv5.阻挠 发表于 2025-3-28 13:05:23
http://reply.papertrans.cn/47/4615/461478/461478_40.png