Amir Habibian

Amir Habibian

Research Scientist

Qualcomm AI Research

I am a research scientist interested in the computational efficiency of neural networks.

At Qualcomm, I have been leading teams developing mobile architectures for:

  • Image and video generation
  • Video perception
  • Neural video compression

I did a PhD on learning multimodal video representations at the University of Amsterdam.

News

Mobile Video Diffusion demo at NeurIPS 2024
We demonstrated our first-in-industry diffusion based video editing running on a phone at NeurIPS
Dec 10, 2024 0 min read
Mobile Video Diffusion demo at NeurIPS 2024
Driving Scene Editing demo at NeurIPS 2024
We demonstrate using dynamic Gaussian splatting for 3D reconstruction of driving scenes to remove, translate or add new vehicles.
Dec 9, 2024 0 min read
Driving Scene Editing demo at NeurIPS 2024
Organizing eLVM Workshop at CVPR 2024
We organize the eLVM: Efficient Large Vision Models workshop at CVPR targeting computational efficiency of training and inference for foundation models. More details including call for papers available at: Website
Jan 24, 2024 0 min read
Data Generation demo at CES 2024
We demonstrated using diffusion models to generate training and test data for perception models in Automotive.
Jan 11, 2024 0 min read
Data Generation demo at CES 2024
Co-organizing RCV Workshop at ICCV 2023
We organize the Resource Efficient Deep Learning for Computer Vision workshop with many interesting posters and keynote talks about recent progress on computational efficiency of computer vision models. Website
Oct 5, 2023 0 min read
See all posts

Publications

Quickly discover relevant content by filtering publications.
Mobile Video Diffusion
Video diffusion models have achieved impressive realism and controllability but are limited by high computational demands, restricting …
Haitam Ben Yahia, Denis Korzhenkov, Ioannis Lelekas, Amir Ghodrati, Amir Habibian
PDF Cite Code Project
Mobile Video Diffusion
MoViE: Mobile Diffusion for Video Editing
Recent progress in diffusion-based video editing has shown remarkable potential for practical applications. However, these methods …
Adil Karjauv, Noor Fathima, Ioannis Lelekas, Fatih Porikli, Amir Ghodrati, Amir Habibian
PDF Cite Code Project
MoViE: Mobile Diffusion for Video Editing
Generative Location Modeling for Spatially Aware Object Insertion
Generative models have become a powerful tool for image editing tasks, including object insertion. However, these methods often lack …
Jooyeol Yun, Davide Abati, Mohamed Omran, Jaegul Choo, Amir Habibian, Auke Wiggers
PDF Cite Project
Generative Location Modeling for Spatially Aware Object Insertion
Object-Centric Diffusion for Efficient Video Editing
Diffusion-based video editing have reached impressive quality and can transform either the global style, local structure, and …
Kumara Kahatapitiya, Adil Karjauv, Davide Abati, Fatih Porikli, Yuki Asano, Amir Habibian
PDF Cite Project
Object-Centric Diffusion for Efficient Video Editing
Clockwork Diffusion: Efficient Generation With Model-Step Distillation
This work aims to improve the efficiency of text-to-image diffusion models. While diffusion models use computationally expensive …
Amir Habibian, Amir Ghodrati, Noor Fathima, Guillaume Sautiere, Risheek Garrepalli, Fatih Porikli, Jens Petersen
PDF Cite Code Project
Clockwork Diffusion: Efficient Generation With Model-Step Distillation
See all publications

Talks

Webinar on Efficient Generation Vision
Presenting our latest research on efficient generation of images, videos, and 3D data. More info
Feb 6, 2024 12:00 AM
Video
Invited talk in the 2nd Workshop on Efficient ML in Vienna
Presenting our latest research on efficient video processing. More info
Jun 13, 2022 12:00 AM
Slides
TIWML AI Podcast on Efficient Video Processing
A podcast with Sam Charlington about our latest research on efficient video processing
Jun 28, 2021 12:00 AM
Video
TIWML AI Podcast on Efficient Video Processing