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

Recently simulation methods have been developed for optical tactile sensorsto enable the Sim2Real learning, i.e., firstly training models in simulationbefore deploying them on the real robot. However, some artefacts in the realobjects are unpredictable, such as imperfections caused by fabricationprocesses, or scratches by the natural wear and tear, and thus cannot berepresented in the simulation, resulting in a significant gap between thesimulated and real tactile images. To address this Sim2Real gap, we propose anovel texture generation network that maps the simulated images intophotorealistic tactile images that resemble a real sensor contacting a realimperfect object. Each simulated tactile image is first divided into two typesof regions: areas that are in contact with the object and areas that are not.The former is applied with generated textures learned from real textures in thereal tactile images, whereas the latter maintains its appearance as when thesensor is not in contact with any object. This makes sure that the artefactsare only applied to the deformed regions of the sensor. Our extensiveexperiments show that the proposed texture generation network can generatethese realistic artefacts on the deformed regions of the sensor, while avoidingleaking the textures into areas of no contact. Quantitative experiments furtherreveal that when using the adapted images generated by our proposed network fora Sim2Real classification task, the drop in accuracy caused by the Sim2Real gapis reduced from 38.43% to merely 0.81%. As such, this work has the potential toaccelerate the Sim2Real learning for robotic tasks requiring tactile sensing.

Reducing Tactile Sim2Real Domain Gaps via Deep Texture Generation Networks