Generative Modeling

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
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
Adversarial Distortion for Learned Video Compression
In this paper, we present a novel adversarial lossy video compression model. At extremely low bit-rates, standard video coding schemes …
Vijay Veerabadran, Reza Pourreza, Amir Habibian, Taco S Cohen
PDF Cite
Adversarial Distortion for Learned Video Compression
Learning Variations in Human Motion via Mix-and-Match Perturbation
Human motion prediction is a stochastic process: Given an observed sequence of poses, multiple future motions are plausible. Existing …
Mohammad Sadegh Aliakbarian, Fatemeh Sadat Saleh, Mathieu Salzmann, Lars Petersson, Stephen Gould, Amir Habibian
PDF Cite Code
Learning Variations in Human Motion via Mix-and-Match Perturbation
Video Compression with Rate-Distortion Autoencoders
In this paper we present a a deep generative model for lossy video compression. We employ a model that consists of a 3D autoencoder …
Amir Habibian, Ties van Rozendaal, Jakub M Tomczak, Taco S Cohen
PDF Cite
Video Compression with Rate-Distortion Autoencoders