Computer Vision & Generative Models

Kian
Izadpanah

Graduate student in Computer Science at the University of Cyprus, conducting research in Generative Models and Computer Vision in collaboration with the Gruvi Lab @ SFU.

UCY — CS Gruvi Lab @ SFU Computer Vision Generative Models
Kian Izadpanah

Nicosia, Cyprus

Research Profile

I am Kian Izadpanah, a Computer Vision researcher specialising in how machines understand and manipulate visual content. I completed my BSc in Computer Engineering at Sharif University of Technology in September 2025, during which I pursued research across several international laboratories.

I am currently pursuing graduate studies in Computer Science at the University of Cyprus (UCY), with an active research collaboration at the Gruvi Lab, Simon Fraser University (SFU). My primary research focus is on Generative Models, with an emphasis on image synthesis and controlled editing.

My research is supervised by Ali Mahdavi-Amiri and Daniel Cohen-Or at SFU, and by Yiorgos Chrysantou and Andreas Aristidou at UCY.

Location
Nicosia, Cyprus
Focus
Generative Models

Computer Vision

Representation learning and visual scene understanding from image and video data.

Generative Models

Diffusion models and flow-based architectures for image synthesis and controlled editing.

Deep Learning

Neural architectures, optimisation methods, and large-scale model training.

Academic Background

Current

University of Cyprus (UCY)

Computer Science
Feb. 2026 — Present

Pursuing graduate research in generative models and computer vision, with a focus on image synthesis, editing, and deep generative architectures.

Supervisors: Y. Chrysantou & A. Aristidou
Collaboration

Gruvi Lab @ Simon Fraser University

Research Collaboration
Feb. 2026 — Present

Active research collaboration on image editing and generative model research, working alongside leading researchers in computer vision and computational graphics.

Supervisors: A. Mahdavi-Amiri & D. Cohen-Or
Completed — Sep. 2025
Sharif University

Sharif University of Technology

BSc in Computer Engineering
Sep. 2021 — Sep. 2025

Completed the undergraduate programme in Computer Engineering with a strong focus on machine learning and computer vision. Enrolled in graduate-level courses in Machine Learning and 3D Computer Vision alongside the core curriculum.

3D Computer Vision ★ Machine Learning ★ Artificial Intelligence Linear Algebra Probability & Statistics Computer Simulation Game Theory Automata Theory Database Design

★ Graduate-level course

Publications

Peer-reviewed and preprint contributions in computer vision and multimedia systems.

2026
Computer Vision · Robotics · Dataset
JRDB-Pose3D: A Multi-person 3D Human Pose and Shape Estimation Dataset for Robotics
Sandika Biswas, Kian Izadpanah, Hamid Rezatofighi
3D Human Pose SMPLX Robotics Dataset
2026
Autonomous Vehicles · VANETs · Multimedia
CITRON: Collaborative Image Stitching and Redundancy Elimination for Energy-Efficient Multimedia Offloading in VANETs
Kian Izadpanah, Mohammad Hossein Soheilian, Bardia Safaei
Image Stitching VANETs Energy Efficiency Multimedia

Research Experience

Research positions and laboratory affiliations.

Jul. 2024 — Dec. 2024
Research Assistant (Summer Internship)
  • Investigated methods for generating and editing 3D human poses, addressing challenges in multi-joint articulation and spatial consistency.
  • Implemented and evaluated 3D body frameworks including Joint2Human, FOF, SMPLX, STAR, and SPIN.
  • Applied text-driven generation models including CLIP, CLIP-PAE, and CLIP-Forge to guide pose synthesis.
Nov. 2023 — Jun. 2024
Research Assistant
  • Contributed to the JRDB dataset, with a focus on improving 3D human pose estimation and annotation quality at scale.
  • Utilised SMPLX for 3D mesh generation and quaternion representations for rotation parameterisation.
Jul. 2023 — Dec. 2023
Research Assistant
  • Investigated cross-domain generalisation for medical object detection, addressing distribution shift in domain adaptation across imaging devices and clinical sites.
  • Proposed a pipeline combining diffusion-based data augmentation with a VAE-based out-of-distribution detector to improve RetinaNet performance on the MIDOG benchmark.
Mar. 2023 — Jun. 2023
Research Assistant
Supervisor: Dr. Hossein Mobahi
  • Developed AGSAM (Adaptive-Gradient SAM), an extension of Sharpness-Aware Minimisation that adapts perturbation magnitude to the local geometry of the loss landscape.
  • Extended the optimisation formulation to incorporate both first-order gradient information and second-order curvature of the objective function.
Feb. 2023 — Apr. 2023
Research Assistant
  • Studied anomaly detection and backdoor attack mechanisms, with an emphasis on detection and defence strategies for deep neural networks.
  • Applied out-of-distribution scoring techniques to video surveillance scenarios.

Technical Skills

Programming Languages

Python Java Go C JavaScript

ML / CV Frameworks

PyTorch TensorFlow NumPy OpenCV

Research Expertise

Generative AI Image Editing Image Generation Interactive Editing Flux & Diffusion Models

Tooling & DevOps

Git GitHub Jupyter

Get in Touch

Open to discussions on research, potential collaborations, and academic opportunities. Please reach out via any of the channels below.