Home

Awesome

HyperE2VID: Improving Event-Based Video Reconstruction via Hypernetworks

Hugging Face Spaces arxiv.org PWC PWC

This is the official repository of our IEEE TIP paper HyperE2VID: Improving Event-Based Video Reconstruction via Hypernetworks by Burak Ercan, Onur Eker, Canberk Sağlam, Aykut Erdem, and Erkut Erdem.

<div align="center"> <a href="https://www.youtube.com/watch?v=BWEV56-E0mE"><img src="media/video_thumbnail.png" alt="HyperE2VID: Improving Event-Based Video Reconstruction via Hypernetworks" width="600"></a> </div

In this work we present HyperE2VID, a dynamic neural network architecture for event-based video reconstruction. Our approach extends existing static architectures by using hypernetworks and dynamic convolutions to generate per-pixel adaptive filters guided by a context fusion module that combines information from event voxel grids and previously reconstructed intensity images. We show that this dynamic architecture can generate higher-quality videos than previous state-of-the-art, while also reducing memory consumption and inference time.

Overview of our proposed HyperE2VID architecture

Citations

If you use code in this repo in an academic context, please cite the following:

@article{ercan2024hypere2vid,
title={{HyperE2VID}: Improving Event-Based Video Reconstruction via Hypernetworks},
author={Ercan, Burak and Eker, Onur and Saglam, Canberk and Erdem, Aykut and Erdem, Erkut},
journal={IEEE Transactions on Image Processing},
year={2024},
volume={33},
pages={1826--1837},
doi={10.1109/TIP.2024.3372460},
publisher={IEEE}}

Acknowledgements