I am an ELLIS PhD student in the Computer Vision Lab at the University of Amsterdam, advised by Prof. Dimitris Tzionas. My research focus on 3D Human Object Interaction (HOI) synthesis, while I am also interested in reconstructing 4D HOIs from videos. Before joining UvA I had the great opportunity to spend 4 months as a research intern at Simon Fraser University working together with Prof. Manolis Savva. Prior to that I completed my Master at the University of Patras collaborating with Prof. Emmanouil Psarakis, while also working as a Lead Quality Assurance Enginner at Hellenic Air Force. I am also a passionate windsurfer. However, when the sea and wind are not there I enjoy spending my time running or going to the gym.
Publications
Synthesizing 3D whole-bodies that realistically grasp objects is useful for animation, mixed reality,
and robotics. This is challenging, because the hands and body need to look natural w.r.t. each other,
the grasped object, as well as the local scene (i.e., a receptacle supporting the object). Only recent
work tackles this, with a divide-and-conquer approach; it first generates a "guiding" right-hand grasp,
and then searches for bodies that match this. However, the guiding-hand synthesis lacks controllability
and receptacle awareness, so it likely has an implausible direction (i.e., a body can't match this without
penetrating the receptacle) and needs corrections through major post-processing. Moreover, the body search
needs exhaustive sampling and is expensive. These are strong limitations. We tackle these with a novel
method called CWGrasp. Our key idea is that performing geometry-based reasoning "early on," instead of
"too late," provides rich "control" signals for inference. To this end, CWGrasp first samples a plausible
reaching-direction vector (used later for both the arm and hand) from a probabilistic model built via
raycasting from the object and collision checking. Then, it generates a reaching body with a desired arm
direction, as well as a "guiding" grasping hand with a desired palm direction that complies with the arm's
one. Eventually, CWGrasp refines the body to match the "guiding" hand, while plausibly contacting the scene.
Notably, generating already-compatible "parts" greatly simplifies the "whole." Moreover, CWGrasp uniquely
tackles both right- and left-hand grasps. We evaluate on the GRAB and ReplicaGrasp datasets. CWGrasp
outperforms baselines, at lower runtime and budget, while all components help performance.
@InProceedings{paschalidis20243d,
title={3D Whole-body Grasp Synthesis with Directional Controllability},
author={Paschalidis, Georgios and Wilschut, Romana and Anti{\'c}, Dimitrije and Taheri, Omid and Tzionas, Dimitrios},
booktitle = {International Conference on 3D Vision (3DV)},
year={2025}
}