Scholar
Alberta Longhini
Google Scholar ID: gwFVvsQAAAAJ
PhD, KTH Royal Institute of Technology
Citations & Impact
Citations
124
H-index
7
i10-index
6
Publications
18
Co-authors
15
list available
Publications
Resume (English only)
Academic Achievements
- Published 'Cloth-Splatting:3D Cloth State Estimation from RGB Supervision' introducing a method for estimating 3D states of cloth from RGB images through a prediction-update framework; Participated in writing a review article 'Unfolding the Literature: A Review of Robotic Cloth Manipulation' providing an overview of the field's development status.
Research Experience
- Currently an intern at Naver Labs Europe, working with the Spatial AI team; PhD Representative - Member of the EECS PhD school council 2023 - 2025; Co-Organizer of ICRA 2024 workshop on Representing and Manipulating Deformable Objects; Co-Organizer of ICRA 2025 workshop on Reflections on Representations and Manipulating Deformable Objects; Co-Organizer of SWOMO RSS 2025 workshop on Structured World Models for Robotic Manipulation; Teaching Assistant for Image Analysis and Computer Vision (DD2423) and Introduction to Robotics (DD2410) from 2022-2024.
Education
- Currently a fourth-year doctoral student at KTH Royal Institute of Technology under the supervision of Professor Danica Kragic.
Background
- Research interests include advancing Physical AI by integrating structured world models and multimodal perception to enable robots to perform complex tasks in unstructured, dynamic environments. Particularly interested in leveraging the interplay between perception and modeling to learn representations of physical properties of real-world objects, embedding physics-based priors into learning systems, and exploring generative and hybrid approaches to improve the generalization and sample efficiency of current Physical AI. Vision is to develop intelligent robotic systems capable of long-horizon autonomy and robust assistive capabilities to support humans in real-world scenarios.
