Abstract
Oblique photogrammetry-based three-dimensional (3D) urban models are widely used for smart cities. In 3D urban models, road signs are small but provide valuable information for navigation. However, due to the problems of sliced shape features, blurred texture and high incline angles, road signs cannot be fully reconstructed in oblique photogrammetry, even with state-of-the-art algorithms. The poor reconstruction of road signs commonly leads to less informative guidance and unsatisfactory visual appearance. In this paper, we present a pipeline for embedding road sign models based on deep convolutional neural networks (CNNs). First, we present an end-to-end balanced-learning framework for small object detection that takes advantage of the region-based CNN and a data synthesis strategy. Second, under the geometric constraints placed by the bounding boxes, we use the scale-invariant feature transform (SIFT) to extract the corresponding points on the road signs. Third, we obtain the coarse location of a single road sign by triangulating the corresponding points and refine the location via outlier removal. Least-squares fitting is then applied to the refined point cloud to fit a plane for orientation prediction. Finally, we replace the road signs with computer-aided design models in the 3D urban scene with the predicted location and orientation. The experimental results show that the proposed method achieves a high mAP in road sign detection and produces visually plausible embedded results, which demonstrates its effectiveness for road sign modeling in oblique photogrammetry-based 3D scene reconstruction.
Highlights
Real-world three-dimensional (3D) urban models are important in building “smart cities” and supporting numerous applications such as city planning, space management, and intelligent traffic systems [1]
Compared to that of faster R-convolutional neural networks (CNNs), the proposed method for synthesizing data improves the Mean average precision (mAP) from 83.2% to 89.7%
We presented a pipeline for embedding road sign models based on deep convolutional neural networks (CNNs)
Summary
Real-world three-dimensional (3D) urban models are important in building “smart cities” and supporting numerous applications such as city planning, space management, and intelligent traffic systems [1]. With the development of unmanned aerial vehicles (UAVs), oblique photogrammetry has been widely used to create 3D urban models because it collects abundant information on a large scale with the benefit of low cost and high efficiency [2,3]. Due to slice- and pole-like shape features, weak texture and high camera incline angle, oblique photogrammetry-based 3D modeling of some artifacts, such as light poles and road signs, remains challenging. Road signs, which play crucial roles in city infrastructure, are set up at the sides of roads and artificially designed with striking colors and regular sliced shapes to provide navigation information and warnings to drivers and pedestrians [4,5]. Reconstructed road signs are fragmentary with discontinuous surfaces and blurred textures in the 3D scene, exposing the defects of oblique photogrammetrybased methods.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
