Nikos Athanasiou

Profile Picture

Ph.D. student at the Max Planck Institute for Intelligent Systems, Perceiving Systems department. My Ph.D. research focuses on exploring and analyzing the underlying causes behind human movement, combining language with human movement to generate and control 3D human motions. I am lucky to be supervised by Michael J. Black and Gül Varol.

Before that I did my BS & MEng. (Diploma) at ECE, NTUA in Athens. Back then, I did research in Natural Language Representations at Speech and Language Processing group under the supervision of Alexandros Potamianos. I also worked for a while at Deeplab, during its early stages, with a wonderful team of people.

Latest News

Research

Supervised by Michael J. Black and Gül Varol, my research aims to understand the intricate details of human movement and generate fine-grained 3D human motions.

I am interested in 3D humans, their interactions with other humans and their environment. I like working with videos, and I would ideally like using all kinds of possible data to train big models. I am also really interested in exploiting large VLMs and LLMs in smart ways to synthesize data and build on top of them.

Publications

MotionFix: Text-based 3D Human Motion Editing

N Athanasiou, A Cseke, M Diomataris, MJ Black, G Varol

SIGGRAPH Asia 2024

WANDR: Intention-guided Human Motion Generation

M Diomataris, N Athanasiou, O Taheri, X Wang, O Hilliges, MJ Black

CVPR 2024

Emotional speech-driven 3D body animation via disentangled latent diffusion

K Chhatre, R Daněček, N Athanasiou, G Becherini, C Peters, MJ Black, T Bolkart

CVPR 2024

TEACH: Temporal Action Composition for 3D Humans

N Athanasiou, M Petrovich, MJ Black, G Varol

3DV 2022

BABEL: Bodies, action and behavior with English labels

AR Punnakkal, A Chandrasekaran, N Athanasiou, A Quiros-Ramirez, MJ Black

CVPR 2021

Ntua-slp at Semeval-2018 task 3: Tracking Ironic Tweets using Ensembles of Word and Character Level Attentive Rnns

C Baziotis, N Athanasiou, P Papalampidi, A Kolovou, G Paraskevopoulos

arXiv preprint arXiv:1804.06659

Ntua-slp at semeval-2018 task 2: Predicting emojis using rnns with context-aware attention

C Baziotis, N Athanasiou, G Paraskevopoulos, N Ellinas, A Kolovou

arXiv preprint arXiv:1804.06657

Contact Me

You can reach me via email at . I strongly prefer email. I am rarely active and almost never keep up with messages in X/Twitter 🐥 or Linkedin 💼. However, I try to reply to all my emails and will probably solve your issues 👾.

Except if you have an issue related to conda. This you have to solve it on your own or seek for professional help. The last two sentences are a joke.