Garvita Tiwari

I am a Ph.D. student at the Real Virtual Humans group, Max Planck Institute for Informatics (MPI-INF), Saarbrücken and Tuebinge AI Center, University of Tuebingen , under the supervision of Prof. Dr. Gerard Pons-Moll.

I completed my Master's in Visual Computing from Saarland University in November 2019. I earned my Bachelor's degree in Electronics and Communication Engineering from IIIT-Allahabad in 2016. Before starting my Master's, I worked as a Software Engineer at Infurnia Technologies in Bengaluru, India.

My research interests include 3D human modeling, clothing models and dynamics, 3D representation and scene understaing, computer vision for graphics, and geometric deep learning.

Email  /  CV  /  Scholar  /  Twitter  /  Github

profile photo

News

  • NRDF was featured as CVPR'24 highlights in Computer Vision News magazine
  • NRDF : Neural Riemannian Distance Fields for Learning Articulated Pose Priors accepted at CVPR 2024
  • CloSe : A 3D Clothing Segmentation Dataset and Model accepted at 3DV 2024
  • Bosch AI Talk Series, April 2023 Neural Distance Fields for Human-Clothing Models and Pose Manifolds
  • Talk @ JHU, April 2023 Pose-NDF: Modeling Human Pose Manifolds with Neural Distance Fields
  • Pose-NDF was featured as Best of ECCV 2022 in Computer Vision News magazine
  • Best Paper Honourable Mention, ECCV 2022 Pose-NDF: Modeling Human Pose Manifolds with Neural Distance Fields
  • PoseNDF accepted at ECCV 2022
  • Talk @ Google Research, July 2022 Pose-NDF: Modeling Human Pose Manifolds with Neural Distance Fields
  • Internship @ Meta July 2022 FAIR, Meta, London
  • Adobe Fellowship Finalist-2021

Publications

NRDF: Neural Riemannian Distance Fields for Learning Articulated Pose Priors
Yannan He, Garvita Tiwari, Tolga Birdal, Jan Eric Lenssen, Gerard Pons-Moll
CVPR, 2024
project page / arXiv / video / code / bibtex

Neural Riemannian Distance Fields (NRDFs), a principled method to learn data-driven priors as subspace of high-dimensional Riemannian manifolds.

CloSe: A 3D Clothing Segmentation Dataset and Model
Dimitrije Antić, Garvita Tiwari, Batuhan Ozcomlekci, Riccardo Marin, Gerard Pons-Moll
3DV, 2024
project page / arXiv / code / bibtex

We present first model that predicts fine-grained clothing segmentation directly from 3D colored point clouds.

Pose-NDF: Modeling Human Pose Manifolds with Neural Distance Fields
Garvita Tiwari, Dimitrije Antić, Jan Eric Lenssen, Nikolaos Sarafianos , Tony Tung, Gerard Pons-Moll
ECCV, 2022   Best Paper Honourable Mention
project page / arXiv / code / bibtex

Pose-NDF learns a manifold of plausible poses as the zero level set of a neural implicit function, extending the idea of modeling implicit surfaces in 3D to the high-dimensional domain SO(3)^K, where a human pose is defined by a single data point, represented by K quaternions.

Neural-GIF: Neural Generalized Implicit Functions for Animating People in Clothing
Garvita Tiwari, Nikolaos Sarafianos , Tony Tung, Gerard Pons-Moll
ICCV, 2021
project page / arXiv / code / bibtex

We present Neural Generalized Implicit Functions (Neural-GIF), to animate people in clothing as a function of body pose. Neural-GIF learns directly from scans, models complex clothing and produces pose-dependent details for realistic animation.

SIZER: A Dataset and Model for Parsing 3D Clothing and Learning Size Sensitive 3D Clothing
Garvita Tiwari, Bharat Lal Bhatnagar , Tony Tung, Gerard Pons-Moll
ECCV, 2020   Oral Presentation
project page / arXiv / code / Data / bibtex

We introduce the SIZER dataset of clothing size variation which includes 100 different subjects wearing casual clothing items in various sizes, totaling to approximately 2000 scans.

Multi-Garment Net: Learning to Dress 3D People from Images
Bharat Lal Bhatnagar , Garvita Tiwari, Christian Theobalt Gerard Pons-Moll
ICCV, 2019
project page / arXiv / code / bibtex

Multi-Garment Network (MGN), a method to predict body shape and clothing, layered on top of the SMPL model from a few frames (1-8) of a video.

Academic Services

Reviewing
(Conferences)

CVPR(2024, 2032, 2022- Outstanding Reviewer), ICCV (2023,2021), ECCV(2024, 2022), 3DV (2024, 2022, 2021, 2020), Eurographics (2021), SIGGRAPH (2021), SIGGRAPH Asia (2024), UIST (2020), MVA (2021)

Reviewing
(Journals)

IEEE Transactions on Visualization and Computer Graphics(2021), Computers & Graphics(2022, 2020), Journal of Systems & Applications in Computer Graphics(2021), PeerJ Computer Science(2021)


Source: Jon Barron's webpage