MFMohsen Fayyaz
Publications

All publications

A selection of research across multimodal learning, video understanding, efficient model design, and machine learning systems.

Research

Publications

For the complete and most up-to-date record, visit Google Scholar.

Google Scholar ↗
2026
STRIVE: Structured Spatiotemporal Exploration for Reinforcement Learning in Video Question Answering
arXiv preprint
Emad Bahrami, Olga Zatsarynna, Parth Pathak, Sunando Sengupta, Jürgen Gall, Mohsen Fayyaz
Paper ↗
2025
Enhancing Temporal Understanding in Video-LLMs through Stacked Temporal Attention in Vision Encoders
NeurIPS
Ali Rasekh, Erfan Bagheri Soula, Omid Daliran, Simon Gottschalk, Mohsen Fayyaz
Paper ↗Project ↗
2025
DeltaLLM: Compress LLMs with Low-Rank Deltas between Shared Weights
arXiv preprint
Liana Mikaelyan, Ayyoob Imani, Mathew Salvaris, Parth Pathak, Mohsen Fayyaz
Paper ↗
2024
Occlusion Handling in 3D Human Pose Estimation with Perturbed Positional Encoding
ECCV
Niloofar Azizi, Mohsen Fayyaz, Horst Bischof
2022
TaylorSwiftNet: Taylor Driven Temporal Modeling for Swift Future Frame Prediction
BMVC
Saber Pourheydari*, Emad Bahrami*, Mohsen Fayyaz*, Gianpiero Francesca, Mehdi Noroozi, Jürgen Gall
* Equal contribution
Paper ↗
2022
Adaptive Token Sampling for Efficient Vision Transformers
ECCV — Oral Presentation (Top 3%)
Mohsen Fayyaz*, Soroush Abbasi Koohpayegani*, Farnoush Rezaei Jafari*, Sunando Sengupta, Hamid Reza Vaezi Joze, Eric Sommerlade, Hamed Pirsiavash, Jürgen Gall
* Equal contribution
Paper ↗Project ↗Code ↗
2022
Fast Weakly Supervised Action Segmentation Using Mutual Consistency
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
Yaser Souri*, Mohsen Fayyaz*, Luca Minciullo, Gianpiero Francesca, Jürgen Gall
* Equal contribution
Paper ↗Code ↗
2021
3D CNNs with Adaptive Temporal Feature Resolutions
CVPR
Mohsen Fayyaz*, Emad Bahrami*, Ali Diba, Mehdi Noroozi, Ehsan Adeli, Luc Van Gool, Jürgen Gall
* Equal contribution
Paper ↗Project ↗
2021
Long Short View Feature Decomposition via Contrastive Video Representation Learning
ICCV
Nadine Behrmann, Mohsen Fayyaz, Jürgen Gall, Mehdi Noroozi
Paper ↗
2020
Large Scale Holistic Video Understanding
ECCV — Spotlight (Top 3%)
Ali Diba*†, Mohsen Fayyaz*†, Vivek Sharma*†, Manohar Paluri, Jürgen Gall, Rainer Stiefelhagen, Luc Van Gool
* Equal contribution · † Listed in alphabetical order
Paper ↗Project ↗
2020
SCT: Set Constrained Temporal Transformer for Set Supervised Action Segmentation
CVPR
Mohsen Fayyaz, Jürgen Gall
Paper ↗Code ↗
2018
AVID: Adversarial Visual Irregularity Detection
ACCV
Mohammad Sabokrou*, Masoud Pourreza*, Mohsen Fayyaz*, Rahim Entezari, Mahmood Fathy, Jürgen Gall, Ehsan Adeli
* Equal contribution
Paper ↗Code ↗
2018
Spatio-Temporal Channel Correlation Networks for Action Classification
ECCV
Ali Diba*, Mohsen Fayyaz*, Vivek Sharma, M. Mahdi Arzani, Rahman Yousefzadeh, Jürgen Gall, Luc Van Gool
* Equal contribution
Paper ↗
2018
Temporal 3D ConvNets by Temporal Transition Layer
CVPR Workshop on Brave New Ideas in Video Understanding
Ali Diba*, Mohsen Fayyaz*, Vivek Sharma, A. Karami, M. Arzani, Rahman Yousefzadeh, Luc Van Gool
* Equal contribution
Paper ↗
2018
Deep-anomaly: Fully Convolutional Neural Network for Fast Anomaly Detection in Crowded Scenes
Computer Vision and Image Understanding
Mohammad Sabokrou*, Mohsen Fayyaz*, Mahmood Fathy, Zahra Moayed, Reinhard Klette
* Equal contribution
Paper ↗
2018
Towards Principled Design of Deep Convolutional Networks: Introducing SimpNet
arXiv preprint
Seyyed Hossein Hasanpour, Mohammad Rouhani, Mohsen Fayyaz, Mohammad Sabokrou, Ehsan Adeli
Paper ↗Code ↗
2017
Deep-cascade: Cascading 3D Deep Neural Networks for Fast Anomaly Detection and Localization in Crowded Scenes
IEEE Transactions on Image Processing
Mohammad Sabokrou, Mohsen Fayyaz, Mahmood Fathy, Reinhard Klette
Paper ↗
2016
STFCN: Spatio-Temporal Fully Convolutional Neural Network for Semantic Segmentation of Street Scenes
ACCV Workshop
Mohsen Fayyaz, Mohammad Sabokrou, Mohammad Hajizadeh, Mahmood Fathy, Fay Huang, Reinhard Klette
Paper ↗Code ↗