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

Traditional approaches to extrinsic calibration use fiducial markers andlearning-based approaches rely heavily on simulation data. In this work, wepresent a learning-based markerless extrinsic calibration system that uses adepth camera and does not rely on simulation data. We learn models forend-effector (EE) segmentation, single-frame rotation prediction and keypointdetection, from automatically generated real-world data. We use atransformation trick to get EE pose estimates from rotation predictions and amatching algorithm to get EE pose estimates from keypoint predictions. Wefurther utilize the iterative closest point algorithm, multiple-frames,filtering and outlier detection to increase calibration robustness. Ourevaluations with training data from multiple camera poses and test data frompreviously unseen poses give sub-centimeter and sub-deciradian averagecalibration and pose estimation errors. We also show that a carefully selectedsingle training pose gives comparable results.

Learning Markerless Robot-Depth Camera Calibration and End-Effector Pose Estimation