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

Self-supervised and language-supervised image models contain rich knowledgeof the world that is important for generalization. Many robotic tasks, however,require a detailed understanding of 3D geometry, which is often lacking in 2Dimage features. This work bridges this 2D-to-3D gap for robotic manipulation byleveraging distilled feature fields to combine accurate 3D geometry with richsemantics from 2D foundation models. We present a few-shot learning method for6-DOF grasping and placing that harnesses these strong spatial and semanticpriors to achieve in-the-wild generalization to unseen objects. Using featuresdistilled from a vision-language model, CLIP, we present a way to designatenovel objects for manipulation via free-text natural language, and demonstrateits ability to generalize to unseen expressions and novel categories ofobjects.

Distilled Feature Fields Enable Few-Shot Language-Guided Manipulation