Aditya Vora

aditya underscore vora at sfu dot ca

I am a Ph.D. student in Computing Science at Simon Fraser University, working with Prof. Hao Zhang.

Before that, I received my Master’s degree from the Indian Institute of Technology, Gandhinagar, India, where I studied Electrical Engineering with a research focus on Computer Vision and Deep Learning, particularly on visual perception problems such as object segmentation and localization under the supervision of Prof. Shanmuganathan Raman.

Prior to that I did my Undergrad in Electronics Engineerig from Birla Vishwakarma Mahavidylaya.

Research

My research focuses on 3D Computer Vision and Machine Learning. Currently, I am interested in designing 3D and 4D neural representations which can enable functionality (like motion/articulation) and spatial control over exisiting or generated 3D assets from foundation models. Some of my papers are highlighted. For complete list of papers, please check my Google Scholar.

2026

Articulate That Object Part (ATOP): 3D Part Articulation via Text and Motion Personalization

Aditya Vora, Sauradip Nag, Kai Wang, and Hao Zhang

ACM Transactions on Graphics (TOG), SIGGRAPH, Los Angeles, CA, USA, 2026

Adapting multi-view diffusion model for controllable multi-view video generation enables 3D shape articulation.

DOI arXiv Code Website
Hierarchical Transformers for Unsupervised 3D Shape Abstraction

Aditya Vora, Lily Goli, Andrea Tagliasacchi, and Hao Zhang

International Conference on 3D Vision (3DV), Vancouver, BC, Canada, 2026

Mapping parent-child relationships of convexes into the cross-attention of a hierarchical transformer enables hierarchical 3D convex decomposition.

arXiv Code Website
Advances in 4D Representation: Geometry, Motion, and Interaction

Mingrui Zhao, Sauradip Nag, Kai Wang, Aditya Vora, Guangda Ji, Peter Chun, Ali Mahdavi-Amiri, and Hao Zhang

Computer Graphics Forum, 2026

A comprehensive survey on 4D representation for geometry, motion, and interaction.

arXiv Code Website

2025

ASIA: Adaptive 3D Segmentation using Few Image Annotations

Sai Raj Kishore Perla, Aditya Vora, Sauradip Nag, Ali Mahdavi-Amiri, and Hao Zhang

SIGGRAPH Asia Conference Papers, Hong Kong, 2025

Use image diffusion models for adaptive 3D segmentation using few image annotations.

DOI arXiv Code Website

2023

DiViNeT: 3D reconstruction from disparate views via neural template regularization

Aditya Vora, Akshay Gadi Patil, and Hao Zhang

Neural Information Processing Systems (NeurIPS), New Orleans, LA, USA, 2023

3D Gaussian-based template regularization for 3D reconstruction from sparse and disparate views.

arXiv Code Website