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

Real world data, especially in the domain of robotics, is notoriously costlyto collect. One way to circumvent this can be to leverage the power ofsimulation to produce large amounts of labelled data. However, training modelson simulated images does not readily transfer to real-world ones. Using domainadaptation methods to cross this "reality gap" requires a large amount ofunlabelled real-world data, whilst domain randomization alone can wastemodeling power. In this paper, we present Randomized-to-Canonical AdaptationNetworks (RCANs), a novel approach to crossing the visual reality gap that usesno real-world data. Our method learns to translate randomized rendered imagesinto their equivalent non-randomized, canonical versions. This in turn allowsfor real images to also be translated into canonical sim images. We demonstratethe effectiveness of this sim-to-real approach by training a vision-basedclosed-loop grasping reinforcement learning agent in simulation, and thentransferring it to the real world to attain 70% zero-shot grasp success onunseen objects, a result that almost doubles the success of learning the sametask directly on domain randomization alone. Additionally, by joint finetuningin the real-world with only 5,000 real-world grasps, our method achieves 91%,attaining comparable performance to a state-of-the-art system trained with580,000 real-world grasps, resulting in a reduction of real-world data by morethan 99%.

Sim-to-Real via Sim-to-Sim: Data-efficient Robotic Grasping via Randomized-to-Canonical Adaptation Networks