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

This paper presents a new technique for learning category-level manipulationfrom raw RGB-D videos of task demonstrations, with no manual labels orannotations. Category-level learning aims to acquire skills that can begeneralized to new objects, with geometries and textures that are differentfrom the ones of the objects used in the demonstrations. We address thisproblem by first viewing both grasping and manipulation as special cases oftool use, where a tool object is moved to a sequence of key-poses defined in aframe of reference of a target object. Tool and target objects, along withtheir key-poses, are predicted using a dynamic graph convolutional neuralnetwork that takes as input an automatically segmented depth and color image ofthe entire scene. Empirical results on object manipulation tasks with a realrobotic arm show that the proposed network can efficiently learn from realvisual demonstrations to perform the tasks on novel objects within the samecategory, and outperforms alternative approaches.

Learning Category-Level Manipulation Tasks from Point Clouds with Dynamic Graph CNNs