Phys-Liquid: A Physics-Informed Dataset for Estimating 3D Geometry and Volume of Transparent Deformable Liquids

TL;DR

Phys-Liquid is a comprehensive physics-informed dataset with simulation images and 3D meshes designed to improve the estimation of transparent deformable liquids' geometry and volume for robotic manipulation tasks.

Contribution

The paper introduces Phys-Liquid, a large-scale, physics-informed dataset with diverse scenarios for transparent liquid perception, along with a validation pipeline for accurate 3D reconstruction.

Findings

Enhanced accuracy in liquid volume estimation

Improved 3D shape reconstruction performance

Outperforms existing benchmarks in liquid perception tasks

Abstract

Estimating the geometric and volumetric properties of transparent deformable liquids is challenging due to optical complexities and dynamic surface deformations induced by container movements. Autonomous robots performing precise liquid manipulation tasks, such as dispensing, aspiration, and mixing, must handle containers in ways that inevitably induce these deformations, complicating accurate liquid state assessment. Current datasets lack comprehensive physics-informed simulation data representing realistic liquid behaviors under diverse dynamic scenarios. To bridge this gap, we introduce Phys-Liquid, a physics-informed dataset comprising 97,200 simulation images and corresponding 3D meshes, capturing liquid dynamics across multiple laboratory scenes, lighting conditions, liquid colors, and container rotations. To validate the realism and effectiveness of Phys-Liquid, we propose a four-stage reconstruction and estimation pipeline involving liquid segmentation, multi-view mask generation, 3D mesh reconstruction, and real-world scaling. Experimental results demonstrate improved accuracy and consistency in reconstructing liquid geometry and volume, outperforming existing benchmarks. The dataset and associated validation methods facilitate future advancements in transparent liquid perception tasks. The dataset and code are available at https://dualtransparency.github.io/Phys-Liquid/.