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 deep reinforcement learning method of progressive viewinpainting for colored semantic point cloud scene completion under volumeguidance, achieving high-quality scene reconstruction from only a single RGB-Dimage with severe occlusion. Our approach is end-to-end, consisting of threemodules: 3D scene volume reconstruction, 2D RGB-D and segmentation imageinpainting, and multi-view selection for completion. Given a single RGB-Dimage, our method first predicts its semantic segmentation map and goes throughthe 3D volume branch to obtain a volumetric scene reconstruction as a guide tothe next view inpainting step, which attempts to make up the missinginformation; the third step involves projecting the volume under the same viewof the input, concatenating them to complete the current view RGB-D andsegmentation map, and integrating all RGB-D and segmentation maps into thepoint cloud. Since the occluded areas are unavailable, we resort to a A3Cnetwork to glance around and pick the next best view for large hole completionprogressively until a scene is adequately reconstructed while guaranteeingvalidity. All steps are learned jointly to achieve robust and consistentresults. We perform qualitative and quantitative evaluations with extensiveexperiments on the 3D-FUTURE data, obtaining better results thanstate-of-the-arts.

Point Cloud Scene Completion with Joint Color and Semantic Estimation from Single RGB-D Image