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

The ability to segment unknown objects in depth images has potential toenhance robot skills in grasping and object tracking. Recent computer visionresearch has demonstrated that Mask R-CNN can be trained to segment specificcategories of objects in RGB images when massive hand-labeled datasets areavailable. As generating these datasets is time consuming, we instead trainwith synthetic depth images. Many robots now use depth sensors, and recentresults suggest training on synthetic depth data can transfer successfully tothe real world. We present a method for automated dataset generation andrapidly generate a synthetic training dataset of 50,000 depth images and320,000 object masks using simulated heaps of 3D CAD models. We train a variantof Mask R-CNN with domain randomization on the generated dataset to performcategory-agnostic instance segmentation without any hand-labeled data and weevaluate the trained network, which we refer to as Synthetic Depth (SD) MaskR-CNN, on a set of real, high-resolution depth images of challenging,densely-cluttered bins containing objects with highly-varied geometry. SD MaskR-CNN outperforms point cloud clustering baselines by an absolute 15% inAverage Precision and 20% in Average Recall on COCO benchmarks, and achievesperformance levels similar to a Mask R-CNN trained on a massive, hand-labeledRGB dataset and fine-tuned on real images from the experimental setup. Wedeploy the model in an instance-specific grasping pipeline to demonstrate itsusefulness in a robotics application. Code, the synthetic training dataset, andsupplementary material are available at https://bit.ly/2letCuE.

Segmenting Unknown 3D Objects from Real Depth Images using Mask R-CNN Trained on Synthetic Data