Mahdi Saleh | 3D CV Researcher

About Me

Hello! I’m Mahdi (also Mehdi) Saleh. I am a Senior Applied Scientist at Zillow, working on 3D vision and generative modeling. If you are interested in PhD internships or related opportunities, please reach out—I am happy to connect.

I completed my doctoral studies at the Technical University of Munich under Federico Tombari and Nassir Navab. My research areas include 3D object pose, hierarchical representations, graph neural networks, and vision-language models. I hold a Master’s degree in Biomedical Computing from TUM and a Bachelor’s in Electrical Control Engineering from IUST. In my leisure time, I enjoy computer graphics and design, web development, and game design.

You can find my latest CV here: CV.

Research Interests

3D computer vision and Machine Learning
6D Object pose estimation and 3D reconstruction
Hierarchical representations and Graph Neural Networks
Vision Language Models and Generative Media

News

[2026] Defended my Ph.D. thesis in April, titled Hierarchical Representations for 3D Shape Understanding. The examination committee passed me with distinction; the degree was awarded summa cum laude.
[2026] Paper accepted at CVPR 2026 (ConceptPose project page).
[2025] Joined VSAI as Senior Research Engineer.
[2025] Paper accepted at CVPR 2025 (ESCAPE project page).
[2025] Zillow acquired VSAI; I joined Zillow as a Senior Applied Scientist.
[2024] Paper accepted at ICRA 2024.
[2023] Paper accepted at ICCV 2023.
[2023] Paper accepted at MICCAI 2023 and 1 paper for ShapeMI MICCAI Workshop.
[2023] Paper accepted at CVPR 2023.
[2023] Starting a PhD research internship at Meta Reality Labs, Redmond, WA.
[2022] Won 2 awards at BOP 6D Object Pose 2022 challenge. Details
[2022] 1 paper accepted at IROS 2022, Kyoto, Japan.
[2022] 2 papers accepted at CVPR 2022, New Orleans, USA.
[2021] 1 paper accepted at NeurIPS 2021.
[2021] Starting a PhD research internship at Facebook Reality Labs, CA.
[2021] Recognized as an Outstanding Reviewer at CVPR 2021.
[2020] Accepted Oral paper at 3DV 2020.

Publications

CVPR

ConceptPose: Training-Free Zero-Shot Object Pose Estimation using Concept Vectors

Kuang, Liming; Velikova, Yordanka; Saleh, Mahdi; Zaech, Jan-Nico; Paudel, Danda Pani; Busam, Benjamin

IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2026

PDF Code Project Page

CVPR

ESCAPE: Equivariant Shape Completion via Anchor Point Encoding

Bekci, Burak; Saleh, Mahdi; Tombari, Federico; Navab, Nassir

IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2025

PDF Code Project Page

ICRA

Physics-Encoded Graph Neural Networks for Deformation Prediction under Contact

Saleh, Mahdi; Sommersperger, Michael; Navab, Nassir; Tombari, Federico

IEEE International Conference on Robotics and Automation, 2024

Code Project Page

IROS

CloudAttention: Efficient Multi-Scale Attention Scheme For 3D Point Cloud Learning

Saleh, Mahdi; Wang, Yige; Navab, Nassir; Busam, Benjamin; Tombari, Federico

International Conference on Intelligent Robots and Systems, 2022

PDF

CVPR

Bending Graphs: Hierarchical Shape Matching using Gated Optimal Transport

Saleh, Mahdi; Wu, Shun-Cheng; Cosmo, Luca; Navab, Nassir; Busam, Benjamin; Tombari, Federico

IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022

PDF Code

CVPR

ZebraPose: Coarse to Fine Surface Encoding for 6DoF Object Pose Estimation

Su, Yongzhi; Saleh, Mahdi; Fetzer, Torben; Rambach, Jason; Navab, Nassir; Busam, Benjamin; Stricker, Didier; Tombari, Federico

IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022

PDF Code Winner of the BOP 2022 Challenge

NeurIPS

CoFiNet: Reliable Coarse-to-fine Correspondences for Robust PointCloud Registration

Yu, Hao; Li, Fu; Saleh, Mahdi; Busam, Benjamin; Ilic, Slobodan

Advances in Neural Information Processing Systems 34, 2021

PDF Code

3DV

Graphite: GRAPH-Induced feaTure Extraction for Point Cloud Registration

Saleh, Mahdi; Dehghani, Shervin; Navab, Nassir; Busam, Benjamin; Tombari, Federico

International Conference on 3D Vision (3DV). IEEE, 2020

PDF Code Oral Presentation

Student Projects

If you are looking for an IDP, guided research, or Master’s thesis around these topics, please contact me.

Here is a list of current and previously supervised projects:

Show supervised projects table — semesters, types, topics, and materials

Semester	Type	Topic	Student(s)	Material
2023/24 WS	Master’s Thesis	Self-Supervised 3D Shape Completion By Growing Graphs	B. Bekci
2023 SS	AT3DCV Project	Exploring Unsupervised Multimodal 3D Understanding	I. Unlu, A. Jevtić, O. Aksoy
2023 SS	AT3DCV Project	Learning Deformations with Hand Interactions	G. Ünver, N. Fest, A. Kolushev
2023 SS	Master’s Thesis	Enhanced 3D ultrasound visualization through Unsupervised Segmentation	Oleksandra Tmenova
2022/23 WS	Master’s Thesis	Vertebrae shape completion in ultrasound	M. Gafencu
2022/23 WS	Master’s Thesis	Real Time Camera Relocalization In Dynamic Environment	E. Shishniashvili
2022 SS	DI-LAB Project	Partial 3D Thyroid Registration	A. Baumann, V. Burgin, E. Hoppe, H. Y. Kim	Final Report
2022 SS	NLO Project	Case Studies Nonlinear Optimization: Deformable Shape Matching	B. Wu, L. Lüdtke, B. Lipp, P. Massetani	Course Webpage
2022 SS	AT3DCV Project	RANSAC with GNN	A. R. Bhattarai, E. D. Franco, K. Han
2022 SS	Master’s Thesis	Domain Agnostic Feature Extraction using Graph Neural Networks	R. Srikanth
2021/22 WS	Guided Research	Self-supervised 3D Detection with Scene Flow	M. Herold, A. Srinivas
2021/22 WS	Master’s Thesis	3D Object Detection and Pose Estimation Using Scene Graphs	S. Jeevanandam
2021/22 WS	Guided Research	Exploring point cloud shape completion and upsampling for 3D object detection	E. Abdelhafez
2021 SS	Lab Project	Hierarchical Retrieval in 3D Scans using GNN	A. Bodonhelyi, A. Desousa
2021 SS	Master’s Thesis	PointPerceiver?: Hierarchical point cloud processing with Transformers	Y. Wang
2021 SS	Master’s Thesis	Weakly-Supervised 3D Object Detection by Self-Ensembling	E. Cakir
2020/21 WS	Master’s IDP	3D Object Detection and Flow Estimation in Dynamic Surgical Environments	D. Gokay, E. Simsar
2020/21 WS	Hiwi	3D Pedestrian Detection from LiDAR Point Clouds	Ch. Yeshwanth
2020 SS	Guided Research	3D Object Detection from Multi-scale Point Clouds Features	N. Danis
2020 SS	Practical Project	Tabletop Object Trajectory Augmentation using 3D Point Clouds	A. Mehrfard, I. Chan, R. Fok
2019/20 WS	Master’s IDP	Graph-based Point Cloud Registration for Tool Tracking	Shervin Dehghani
2019/20 WS	Hiwi Project	Visual Assistance with 3D captioning	HyunJun Jung
2018/19 WS	Practical Project	Autonomous Garbage Collector using 3D Computer Vision and Reinforcement Learning	HyunJun Jung, Damian Bogunowicz, Sangram Gupta	Final presentation; Video Welcome.AI
2018/19 WS	Practical Project	3D Reconstruction and Semantic Segmentation for Furniture Replacement	Vasiliki Sideri, Shervin Dehghani, Tam Tran	Final presentation
2018/19 WS	Bachelor’s Thesis	Deep Reinforcement Learning Simulation for Path Planning	Tobias Valinski
2018/19 SS	Lab Course	3D Joint Motion Visualization for Sports or Physical Therapy	Sri Divya Vajapeyajula
2018/19 SS	Lab Course	Live Avatar of 3D Human Pose Tracking	Arturo Diaz Coca
2018 SS	Bachelor’s IDP	Implementation of Mobile Robot Communication Interface and Sensor Control	Krist S.