Microscopic 3D shape reconstruction using depth from focus (DFF) is crucial in precision manufacturing for 3D modeling and quality control. However, the absence of high-precision microscopic DFF datasets and the significant differences between existing DFF datasets and microscopic DFF data in optical design, imaging principles and scene characteristics hinder the performance of current DFF models in microscopic tasks. To address this, we introduce M3D, a novel microscopic DFF dataset, constructed using a self-developed microscopic device. It includes multi-focus image sequences of 1,952 scenes across five categories, with depth labels obtained through the 3D TFT algorithm applied to dense image sequences for initial depth estimation and calibration. All labels are then compared and analyzed against the design values, and those with large errors are eliminated. We also propose M3DNet, a frequency-aware end-to-end network, to tackle challenges like shallow depth-of-field (DoF) and weak textures. Results show that M3D compensates for the limitations of macroscopic DFF datasets and extends DFF applications to microscopic scenarios. M3DNet effectively captures rapid focus decay and improves performance on public DFF datasets by leveraging superior global feature extraction. Additionally, it exhibits strong robustness even in extreme conditions. Dataset and code are available at https://github.com/jiangfeng-Z/M3D.

Abstract Fine‐grained classification is a challenging task due to the difficulty of finding discriminative features and the localization of feature regions. To handle these challenges, a novel visual classification framework on panoramic attention mechanism that combines multiple attention networks to locate and identify features with more semantic interest is proposed. Firstly, based on the classical convolutional neural network, the global information of the image feature is expressed by linear fusion. Secondly, the foreground attention branch is used to further extract the distinguishing details of the salient features. Then, more features are mined from the complementary object area through the background attention branch to learn more perfect fine‐grained feature expression. Finally, three network branches are trained together to enhance the network's ability to express representative features of fine‐grained images. Our model can be viewed as a multi‐branch network, which benefits each other and optimizes the network together. Experiments were conducted on CUB‐200‐2011, Stanford Dogs and FGVC‐Aircraft datasets, and the accuracy was used as the quantitative measurement. Experimental results show that the proposed method has the highest accuracy; the average accuracy is 89.8%. It is effective and superior to the current advanced methods.

In this paper, we propose a framework to assess visual complexity of paintings. This framework provides a machine learning scheme for investigating the relationship between human visual complexity perception and low-level image features. Since the global and local characteristics of paintings affect human's holistic impression and detail perception, we design a set of methods to extract the features that represent the global and local characteristics of paintings. By feature selection, we look into the role that each image feature plays in assessing visual complexity. Then the selected features are combined by a Support Vector Machine for classification. Experimental results indicate that the proposed work can predict the visual complexity perception of paintings with the accuracy of 88.13%, which is highly close to the assessments given by humans. Compared with the conventional measure of complexity, our approach considers human visual perception and performs more efficiently in assessing visual complexity of painting images.