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

How can we segment varying numbers of objects where each specific objectrepresents its own separate class? To make the problem even more realistic, howcan we add and delete classes on the fly without retraining or fine-tuning?This is the case of robotic applications where no datasets of the objects existor application that includes thousands of objects (E.g., in logistics) where itis impossible to train a single model to learn all of the objects. Most currentresearch on object segmentation for robotic grasping focuses on class-levelobject segmentation (E.g., box, cup, bottle), closed sets (specific objects ofa dataset; for example, YCB dataset), or deep learning-based template matching.In this work, we are interested in open sets where the number of classes isunknown, varying, and without pre-knowledge about the objects' types. Weconsider each specific object as its own separate class. Our goal is to developan object detector that requires no fine-tuning and can add any object as aclass just by capturing a few images of the object. Our main idea is to breakthe segmentation pipelines into two steps by combining unseen objectsegmentation networks cascaded by class-adaptive classifiers. We evaluate ourclass-adaptive object detector on unseen datasets and compare it to a trainedMask R-CNN on those datasets. The results show that the performance varies frompractical to unsuitable depending on the environment setup and the objectsbeing handled. The code is available in our DoUnseen library repository.

DoUnseen: Tuning-Free Class-Adaptive Object Detection of Unseen Objects for Robotic Grasping