Deep learning for detecting building façade elements from images considering prior knowledge

Capital Structure and Firm Performance: Evidence From 2021 Best-Performed Stocks in Nigeria

“…(Abdullah & Tursoy, 2021) Support this contribution with MM's idea of no tax in the stock market. As a result, the study is consistent with the recommendations of (Zhang et al, 2022) and has provided academics, business executives, and investors (Mentzer, 2008) with feasible expertise for venture decision-making. (Fatmasari et al, 2021) Agree that stock return represents the value of a firm, suggesting that an increased stock market return indicates an increase in firm performance.…”

“…Studies build or encode elements of thought into existing knowledge (Zhang et al, 2022). Many studies examined the type and strength of the relationship between capital structure and firm performance using earnings before interest and Tax (EBIT) or net income (Hung et al, 2021;Javed et al, 2014;Mushafiq et al, 2021).…”

Cross Field-Based Segmentation and Learning-Based Vectorization for Rectangular Windows

“…In particular, attention modules can be integrated into backbones [24][25] or head networks [26] [27] to encode distant dependencies or heterogeneous interactions, thus boosting the segmentation quality. Zhang et al [28] employ a dual attentional network (DAN) module to model long-range dependencies, and introduce a novel symmetric loss function to encode prior knowledge improving the predictions of fac ¸ade elements. To the best of the authors' knowledge, this work is the actual state-of-the-art for fac ¸ade semantic segmentation.…”

“…The annotation rule follows uniform Haussmanian-style grammar, i.e., all windows are annotated as rectangles, even though some of them are arc-shaped. This dataset has been widely used to evaluate window detection or fac ¸ade segmentation approaches [28], [61], [62], [3].…”

Dual attention deep learning network for automatic steel surface defect segmentation

“…Over the last decade, the rapid growth of labeled images and computational power has triggered a surge of interest in deep learning, which has achieved higher accuracy and reliability in computer‐vision problems. In other words, deep learning is gradually replacing traditional approaches to become the mainstream in automated image processing and visual imagery analysis nowadays (Zhang et al, 2022). Deep learning as a subset of machine learning can be interpreted as neural networks with multiple layers and neurons, which imitates the function of human brains to capture intricate structures in image data for decision‐making (Zhao et al., 2020).…”