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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Majority of the perception methods in robotics require depth informationprovided by RGB-D cameras. However, standard 3D sensors fail to capture depthof transparent objects due to refraction and absorption of light. In thispaper, we introduce a new approach for depth completion of transparent objectsfrom a single RGB-D image. Key to our approach is a local implicit neuralrepresentation built on ray-voxel pairs that allows our method to generalize tounseen objects and achieve fast inference speed. Based on this representation,we present a novel framework that can complete missing depth given noisy RGB-Dinput. We further improve the depth estimation iteratively using aself-correcting refinement model. To train the whole pipeline, we build a largescale synthetic dataset with transparent objects. Experiments demonstrate thatour method performs significantly better than the current state-of-the-artmethods on both synthetic and real world data. In addition, our approachimproves the inference speed by a factor of 20 compared to the previous bestmethod, ClearGrasp. Code and dataset will be released athttps://research.nvidia.com/publication/2021-03_RGB-D-Local-Implicit.

RGB-D Local Implicit Function for Depth Completion of Transparent Objects