Progressive Prioritized Multi-view Stereo

Abstract

This work proposes a progressive patch based multiview stereo algorithm able to deliver a dense point cloud at any time. This enables an immediate feedback on the reconstruction process in a user centric scenario. With increasing processing time, the model is improved in terms of resolution and accuracy. The algorithm explicitly handles input images with varying effective scale and creates visually pleasing point clouds. A priority scheme assures that the limited computational power is invested in scene parts, where the user is most interested in or the overall error can be reduced the most. The architecture of the proposed pipeline allows fast processing times in large scenes using a pure open-source CPU implementation. We show the performance of our algorithm on challenging standard datasets as well as on real-world scenes and compare it to the baseline.

3d Reconstruction With a Collaborative Approach Based on Smartphones and a Cloud-Based Server

“…Two pipelines are under testing: the first, described in Poiesi et al (2017) and Nocerino et al (2017) is based on approaches proposed by Sweeney et al (2015), Schonberger et al (2016), Locher et al (2016a) and Locher et al (2016b). The second procedure, hereafter presented, follows the SfM/DIM pipeline presented by Schonberger and Frahm (2016).…”

Progressive 3D Modeling All the Way

“…The algorithm works purely in batch mode and therefore is neither progressive nor adaptive to changing calibration. The recently published HPMVS [10] can be seen as a progressive version of PMVS. HPMVS can progressively deliver a dense 3D point cloud which gets more detailed the longer the algorithm runs.…”

“…The presented algorithm makes use of an open source algorithm HPMVS [10] and interacts seamlessly with it. In HPMVS, individual 3D points are internally organized in a hierarchical octree structure and a processing loop repeatedly densifies the point cloud in an expansion procedure and increases the resolution in a branching step.…”

“…While incremental SfM can incorporate new images into an existing 3D model and provide an updated output soon after [16], current MVS algorithms cannot handle changes in the input calibration. Even the recently published Progressive Prioritized Multi-view Stereo (HPMVS) algorithm [10], which can deliver a progressive multiscale output, cannot handle changes in input data. As a result, a modification of the sparse SfM model requires the execution of the MVS algorithm from scratch and an already computed dense model is completely invalidated.…”

Towards Gesture-Based Multi-User Interactions In Collaborative Virtual Environments

“…A smartphone App acquires images and uploads them to a reconstruction server that progressively creates 3D models using algorithms such as Structure from Motion (Wu, 2013), Multi-View Stereo (Locher et al, 2016) and Meshing (Kazhdan et al, 2006) (Schonberger and Frahm, 2016). While a user is scanning an object, the App receives real-time feedback about the status of the reconstruction from the server, thus enabling them to focus on parts of the scanned object that deserve more attention.…”