Emad Bahrami

emadbr.jpg

I’m Emad, a PhD student in the Computer Vision Group at the University of Bonn, supervised by Prof. Jürgen Gall. My primary research area is video understanding, currently focused on enhancing temporal modeling in Multimodal LLMs. Previously, I worked on tasks such as temporal action segmentation and long-term dense anticipation. I’m currently a research intern at Microsoft, working on Multimodal Large Language Models (LLMs) for video perception and reasoning.

Before starting my PhD, I was a researcher at Deep MI, specializing in semantic segmentation of human brain MRI scans. Additionally, I spent time as a visiting researcher in the Computer Vision Group, focusing on action recognition and future frame prediction, also under the supervision of Prof. Gall. I completed my bachelor’s degree at the University of Tehran.

news

May 2025 Glad to be recognized as an Outstanding Reviewer at CVPR 2025 🎉
May 2025 I’ve joined Microsoft as a Research Intern.

selected publications [view all]

  1. Towards Generalizing Temporal Action Segmentation to Unseen Views
    Emad Bahrami*, Olga Zatsarynna*, Gianpiero Francesca, and 1 more author
    * Indicates equal contribution.
    arxiv preprint 2025
    PDF
  2. MANTA: Diffusion Mamba for Efficient and Effective Stochastic Long-Term Dense Anticipation
    Olga Zatsarynna, Emad Bahrami, Yazan Abu Farha, and 2 more authors
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2025
    PDF
  3. Gated Temporal Diffusion for Stochastic Long-Term Dense Anticipation
    Olga Zatsarynna*, Emad Bahrami*, Yazan Abu Farha, and 2 more authors
    * Indicates equal contribution.
    European Conference on Computer Vision (ECCV) 2024
    PDF Code
  4. How Much Temporal Long-Term Context is Needed for Action Segmentation?
    Emad Bahrami, Gianpiero Francesca, and Juergen Gall
    IEEE International Conference on Computer Vision (ICCV) 2023
    PDF Code
  5. CerebNet: A fast and reliable deep-learning pipeline for detailed cerebellum sub-segmentation
    Jennifer Faber*, David Kügler*, Emad Bahrami*, and 14 more authors
    * Indicates equal contribution.
    NeuroImage 2022
    HTML Code
  6. Robust Action Segmentation from Timestamp Supervision
    Yaser Souri*, Yazan Abu Farha*, Emad Bahrami*, and 2 more authors
    * Indicates equal contribution.
    British Machine Vision Conference (BMVC) 2022
    HTML PDF Code
  7. TaylorSwiftNet: Taylor Driven Temporal Modeling for Swift Future Frame Prediction
    Saber* Pourheydari, Emad Bahrami*, Mohsen Fayyaz*, and 3 more authors
    * Indicates equal contribution.
    British Machine Vision Conference (BMVC) 2022
    HTML PDF Code
  8. 3D CNNs With Adaptive Temporal Feature Resolutions
    Mohsen Fayyaz*, Emad Bahrami*, Ali Diba, and 4 more authors
    * Indicates equal contribution.
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2021
    HTML PDF Code