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

Liquid perception is critical for robotic pouring tasks. It usually requiresthe robust visual detection of flowing liquid. However, while recent works haveshown promising results in liquid perception, they typically require labeleddata for model training, a process that is both time-consuming and reliant onhuman labor. To this end, this paper proposes a simple yet effective frameworkPourIt!, to serve as a tool for robotic pouring tasks. We design a simple datacollection pipeline that only needs image-level labels to reduce the relianceon tedious pixel-wise annotations. Then, a binary classification model istrained to generate Class Activation Map (CAM) that focuses on the visualdifference between these two kinds of collected data, i.e., the existence ofliquid drop or not. We also devise a feature contrast strategy to improve thequality of the CAM, thus entirely and tightly covering the actual liquidregions. Then, the container pose is further utilized to facilitate the 3Dpoint cloud recovery of the detected liquid region. Finally, theliquid-to-container distance is calculated for visual closed-loop control ofthe physical robot. To validate the effectiveness of our proposed method, wealso contribute a novel dataset for our task and name it PourIt! dataset.Extensive results on this dataset and physical Franka robot have shown theutility and effectiveness of our method in the robotic pouring tasks. Ourdataset, code and pre-trained models will be available on the project page.

PourIt!: Weakly-supervised Liquid Perception from a Single Image for Visual Closed-Loop Robotic Pouring