Abstract
The reconstruction of the 3D geometry of a scene based on image sequences has been a very active field of research for decades. Nevertheless, there are still existing challenges in particular for homogeneous parts of objects. This paper proposes a solution to enhance the 3D reconstruction of weakly-textured surfaces by using standard cameras as well as a standard multi-view stereo pipeline. The underlying idea of the proposed method is based on improving the signal-to-noise ratio in weakly-textured regions while adaptively amplifying the local contrast to make better use of the limited numerical range in 8-bit images. Based on this premise, multiple shots per viewpoint are used to suppress statistically uncorrelated noise and enhance low-contrast texture. By only changing the image acquisition and adding a preprocessing step, a tremendous increase of up to 300% in completeness of the 3D reconstruction is achieved.
Highlights
The reconstruction of the 3D geometry of a scene based on image sequences has been a very active field of research
This paper proposes to enhance the 3D reconstruction of weaklytextured surfaces by using standard cameras as well as a standard multi-view stereo (MVS) pipeline
Weakly-textured surfaces pose a great challenge for MVS reconstructions because the texture is effectively hidden by sensor noise and can no longer reliably be matched between images
Summary
The reconstruction of the 3D geometry of a scene based on image sequences has been a very active field of research. What is usually meant with this phrase is that the existing physical texture is either too fine-grained to be captured by the spatial resolution of a given camera, or that it has insufficient contrast The latter causes the contribution of the physical texture to the measured signal to drop below the contribution of the measurement noise, which makes a distinction of the two close to impossible. All that is necessary to exploit the existing texture in this case is to enhance the signal-to-noiseratio Based on this premise, this paper proposes to use multiple shots per viewpoint to suppress statistically uncorrelated noise and enhance low-contrast texture. This paper proposes to use multiple shots per viewpoint to suppress statistically uncorrelated noise and enhance low-contrast texture It models the measured signal as a mixture of truly random as well as fixed-pattern noise and leads to a significant improvement of image quality and (more importantly) of the completeness of the reconstruction. A flexible and open-source C++ implementation of the proposed framework is publicly available (Ley, 2016)
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 callMore From: ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: 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.
