论文标题
D-VPNET：自然场景中实时占主导地位的消失点检测网络
D-VPnet: A Network for Real-time Dominant Vanishing Point Detection in Natural Scenes
论文作者
论文摘要
作为线性透视的重要组成部分，消失点（VPS）为将对象从2D照片映射到3D空间提供了有用的线索。现有方法主要集中于提取结构特征，例如线条或轮廓，然后聚集这些特征以检测VPS。但是，由于在室外环境中检测到的线段和轮廓大量，这些技术遭受了模棱两可的信息。在本文中，我们提出了一个新的卷积神经网络（CNN），以检测自然场景中的主要VP，即主导的消失点检测网络（D-VPNET）。我们方法的关键组成部分是特征线段提案单元（FLPU），可以直接用于预测主要VP的位置。此外，该模型还使用两个主要平行线作为助手来确定主要VP的位置。使用公共数据集和基于并行线的消失点（PLVP）数据集测试了所提出的方法。实验结果表明，我们方法的检测准确性在各种条件下实时优于最先进方法的检测准确性，达到115fps的速率。
As an important part of linear perspective, vanishing points (VPs) provide useful clues for mapping objects from 2D photos to 3D space. Existing methods are mainly focused on extracting structural features such as lines or contours and then clustering these features to detect VPs. However, these techniques suffer from ambiguous information due to the large number of line segments and contours detected in outdoor environments. In this paper, we present a new convolutional neural network (CNN) to detect dominant VPs in natural scenes, i.e., the Dominant Vanishing Point detection Network (D-VPnet). The key component of our method is the feature line-segment proposal unit (FLPU), which can be directly utilized to predict the location of the dominant VP. Moreover, the model also uses the two main parallel lines as an assistant to determine the position of the dominant VP. The proposed method was tested using a public dataset and a Parallel Line based Vanishing Point (PLVP) dataset. The experimental results suggest that the detection accuracy of our approach outperforms those of state-of-the-art methods under various conditions in real-time, achieving rates of 115fps.
