Dimitrije Antić

Teaching machines to perceive humans at CVLab @ University of Amsterdam

prof_pic.jpg

Hello World, I am a first-year Ph.D. student in Computer Vision at the University of Amsterdam (UvA), in the Computer Vision Lab. My research is about human-centric Computer Vision and AI, mainly human-object interaction in 3D and human body reconstruction. I am advised by Prof. Dimitris Tzionas and Prof. Theo Gevers.

Earlier, I obtained an MSc degree in Machine Learning at the University of Tuebingen. I was fortunate to work in the Real Virtual Humans Group led by Prof. Gerard Pons-Moll. I also spent some time at the Max Planck Institute for Intelligent Systems in the Autonomous Learning Group.

Even earlier, I obtained a BSc degree in Informatics at the Faculty of Mathematics, University of Belgrade, Serbia.

news

Nov 06, 2024 Our paper 3D Whole-body Grasp Synthesis with Directional Controllability just got accepted to 3DV’25.
Sep 24, 2024 Checkout our latest preprint SDFit: 3D Object Pose and Shape by Fitting a Morphable SDF to a Single Image .
Mar 20, 2024 If you need to segment clothing from a 3D scan, checkout our work CloSe: A 3D Clothing Segmentation Dataset and Model presented at 3DV’24. The dataset and model are available at this GitHub repo.
Mar 01, 2023 I have started my PhD advised by Prof. Dimitris Tzionas and Prof. Theo Gevers at the Computer Vision Lab at the University of Amsterdam.

selected publications

  1. close.png
    CloSe: A 3D Clothing Segmentation Dataset and Model
    Dimitrije AnticGarvita Tiwari, Batuhan Ozcomlekci, Riccardo Marin, and Gerard Pons-Moll
    In International Conference on 3D Vision (3DV), Mar 2024 - Poster
  2. pose-ndf.png
    Pose-NDF: Modeling Human Pose Manifolds with Neural Distance Fields
    In European Conference on Computer Vision (ECCV), Oct 2022 - Best Paper Honorable Mention
  3. beta-nll.png
    On the Pitfalls of Heteroscedastic Uncertainty Estimation with Probabilistic Neural Networks
    Maximilian Seitzer, Arash Tavakoli, Dimitrije Antic, and Georg Martius
    In International Conference on Learning Representations, Apr 2022 - Poster