English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present a real-time visual-inertial dense mapping method capable ofperforming incremental 3D mesh reconstruction with high quality using onlysequential monocular images and inertial measurement unit (IMU) readings. 6-DoFcamera poses are estimated by a robust feature-based visual-inertial odometry(VIO), which also generates noisy sparse 3D map points as a by-product. Wepropose a sparse point aided multi-view stereo neural network (SPA-MVSNet) thatcan effectively leverage the informative but noisy sparse points from the VIOsystem. The sparse depth from VIO is firstly completed by a single-view depthcompletion network. This dense depth map, although naturally limited inaccuracy, is then used as a prior to guide our MVS network in the cost volumegeneration and regularization for accurate dense depth prediction. Predicteddepth maps of keyframe images by the MVS network are incrementally fused into aglobal map using TSDF-Fusion. We extensively evaluate both the proposedSPA-MVSNet and the entire visual-inertial dense mapping system on severalpublic datasets as well as our own dataset, demonstrating the system'simpressive generalization capabilities and its ability to deliver high-quality3D mesh reconstruction online. Our proposed dense mapping system achieves a39.7% improvement in F-score over existing systems when evaluated on thechallenging scenarios of the EuRoC dataset.

SimpleMapping: Real-Time Visual-Inertial Dense Mapping with Deep Multi-View Stereo