Shaifali Parashar

I am a CNRS Research scientist at LIRIS, INSA-Lyon. Previously, I was a postdoc researcher at CVLAB, EPFL working on 3D computer vision with Prof. Pascal Fua and Dr. Mathieu Salzmann. I received my PhD in Computer Vision at University of Clermont-Auvergne, France in 2017. I was supervised by Prof. Adrien Bartoli and Dr. Daniel Pizarro. My research interests are non-rigid 3D reconstruction, deformation modeling and geometric shape processing.

Email  /  Scholar  /  Github  /  LinkedIn

• 2024: I will be co-organising TradiCV workshop which will be held in conjunction with ECCV 2024.
• 2024: Best Student Paper award at 3DV 2024.

• 2022-2026: Project RHINO (Funded by ANR-JCJC).

• 2022-2025: Ruochen Chen. PhD student under Project RHINO.
• 2023-2026: Thuy Tran. PhD student under Project RHINO.
• 2024: Blanca Hermosilla. Master student under Project RHINO.

A Closed-Form, Pairwise Solution to Local Non-Rigid Structure-from-Motion
Shaifali Parashar, Yuxuan Long, Mathieu Salzmann, Pascal Fua
TPAMI, 2024  
Code

GAPS: Geometry-Aware, Physics-Based, Self-Supervised Neural Garment Draping
Ruochen Chen, Liming Chen, Shaifali Parashar
3DV, 2024   (Best student paper award)
Code / Video

Temporally-Coherent Surface Reconstruction via Metric-Consistent Atlases
Jan Bednarik, Vladimir G. Kim, Siddhartha Chaudhuri, Shaifali Parashar, Mathieu Salzmann, Pascal Fua, Noam Aigerman
ICCV, 2021  
Code

Robust Isometric Non-Rigid Structure-from-Motion
Shaifali Parashar, Daniel Pizarro and Adrien Bartoli
TPAMI, 2021  
Code

DefSLAM: Tracking and Mapping of Deforming Scenes from Monocular Sequences
Jose Lamarca, Shaifali Parashar, Adrien Bartoli and J. M. M. Montiel
TRO, 2020  
Code

GarNet++: Improving Fast and Accurate Static 3D Cloth Draping by Curvature Loss
Erhan Gundogdu, Victor Constantin, Shaifali Parashar, Amrollah Seifoddini, Minh Dang, Mathieu Salzmann and Pascal Fua
TPAMI, 2020  
Code & Datasets

Shape Reconstruction by Learning Differentiable Surface Representations
Jan Bednarik, Shaifali Parashar, Erhan Gundogdu, Mathieu Salzmann and Pascal Fua
CVPR, 2020  
Code

Local Non-Rigid Structure-from-Motion from Diffeomorphic Mappings
Shaifali Parashar, Mathieu Salzmann and Pascal Fua
CVPR, 2020  
Code

Local Deformable 3D Reconstruction with Cartan's Connections
Shaifali Parashar, Daniel Pizarro and Adrien Bartoli
TPAMI, 2019  
Code

3DVFX: 3D Video Editing using Non-Rigid Structure-from-Motion
Shaifali Parashar and Adrien Bartoli
Eurographics (Short papers), 2019   (Oral presentation)

Self-Calibrating Isometric Non-Rigid Structure-from-Motion
Shaifali Parashar, Adrien Bartoli and Daniel Pizarro
ECCV, 2018  
Code

Isometric Non-Rigid Shape-from-Motion in Linear Time
Shaifali Parashar, Daniel Pizarro and Adrien Bartoli
CVPR, 2016   (Oral presentation)
Isometric Non-Rigid Shape-from-Motion with Riemannian Geometry in Linear Time
Shaifali Parashar, Daniel Pizarro and Adrien Bartoli
TPAMI, 2017  
Code / Video
Lecture and Tutorial on Isometric NRSfM by Paul Gafton during the virtual winter school on deformable SLAM, July 2021.

As-Rigid-As-Possible Volumetric Shape-From-Template
Shaifali Parashar, Daniel Pizarro, Adrien Bartoli and Toby Collins
ICCV, 2015  
Code / Video

Image-based deformable 3D reconstruction using differential geometry and cartan's connections
Shaifali Parashar
Université Clermont-Auvergne, France, 2017  
(Winner of thesis prize 2018)

Computer vision and machine learning conferences (CVPR, ICCV, ECCV, NeurIPS, ICLR, ICML, AAAI, ACCV, 3DV, WACV) and journals (PAMI, IJCV), Robotics conference IROS and journal I-RAL

• June 2024: Invited talk at at ENS Lyon.
• September 2022: Seminar at LIRIS, Lyon on September 29, 14h 30. Virtual participation link.
• March 2022: Invited talk at the University of Stirling, Scotland.
• July 2021: Keynote speech at the virtual winter school on Deformable SLAM hosted by UTS, Australia. Also see the Panel Discussion on future and challenges of SLAM.

• Fall 2022 [M2 ID3D, M2 DISS]: Introduction to Computer Vision.
• Fall 2023 [M2 ID3D, M2 DISS]: Introduction to Computer Vision.