EvFocus: Learning to Reconstruct Sharp Images from Out-of-Focus Event Streams

Lin Zhu, Xiantao Ma, Xiao Wang, Lizhi Wang, Hua Huang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:79963-79984, 2025.

Abstract

Event cameras are innovative sensors that capture brightness changes as asynchronous events rather than traditional intensity frames. These cameras offer substantial advantages over conventional cameras, including high temporal resolution, high dynamic range, and the elimination of motion blur. However, defocus blur, a common image quality degradation resulting from out-of-focus lenses, complicates the challenge of event-based imaging. Due to the unique imaging mechanism of event cameras, existing focusing algorithms struggle to operate efficiently on sparse event data. In this work, we propose EvFocus, a novel architecture designed to reconstruct sharp images from defocus event streams for the first time. Our work includes the development of an event-based out-of-focus camera model and a simulator to generate realistic defocus event streams for robust training and testing. EvDefous integrates a temporal information encoder, a blur-aware two-branch decoder, and a reconstruction and re-defocus module to effectively learn and correct defocus blur. Extensive experiments on both simulated and real-world datasets demonstrate that EvFocus outperforms existing methods across varying lighting conditions and blur sizes, proving its robustness and practical applicability in event-based defocus imaging.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-zhu25n, title = {{E}v{F}ocus: Learning to Reconstruct Sharp Images from Out-of-Focus Event Streams}, author = {Zhu, Lin and Ma, Xiantao and Wang, Xiao and Wang, Lizhi and Huang, Hua}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {79963--79984}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/zhu25n/zhu25n.pdf}, url = {https://proceedings.mlr.press/v267/zhu25n.html}, abstract = {Event cameras are innovative sensors that capture brightness changes as asynchronous events rather than traditional intensity frames. These cameras offer substantial advantages over conventional cameras, including high temporal resolution, high dynamic range, and the elimination of motion blur. However, defocus blur, a common image quality degradation resulting from out-of-focus lenses, complicates the challenge of event-based imaging. Due to the unique imaging mechanism of event cameras, existing focusing algorithms struggle to operate efficiently on sparse event data. In this work, we propose EvFocus, a novel architecture designed to reconstruct sharp images from defocus event streams for the first time. Our work includes the development of an event-based out-of-focus camera model and a simulator to generate realistic defocus event streams for robust training and testing. EvDefous integrates a temporal information encoder, a blur-aware two-branch decoder, and a reconstruction and re-defocus module to effectively learn and correct defocus blur. Extensive experiments on both simulated and real-world datasets demonstrate that EvFocus outperforms existing methods across varying lighting conditions and blur sizes, proving its robustness and practical applicability in event-based defocus imaging.} }
Endnote
%0 Conference Paper %T EvFocus: Learning to Reconstruct Sharp Images from Out-of-Focus Event Streams %A Lin Zhu %A Xiantao Ma %A Xiao Wang %A Lizhi Wang %A Hua Huang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-zhu25n %I PMLR %P 79963--79984 %U https://proceedings.mlr.press/v267/zhu25n.html %V 267 %X Event cameras are innovative sensors that capture brightness changes as asynchronous events rather than traditional intensity frames. These cameras offer substantial advantages over conventional cameras, including high temporal resolution, high dynamic range, and the elimination of motion blur. However, defocus blur, a common image quality degradation resulting from out-of-focus lenses, complicates the challenge of event-based imaging. Due to the unique imaging mechanism of event cameras, existing focusing algorithms struggle to operate efficiently on sparse event data. In this work, we propose EvFocus, a novel architecture designed to reconstruct sharp images from defocus event streams for the first time. Our work includes the development of an event-based out-of-focus camera model and a simulator to generate realistic defocus event streams for robust training and testing. EvDefous integrates a temporal information encoder, a blur-aware two-branch decoder, and a reconstruction and re-defocus module to effectively learn and correct defocus blur. Extensive experiments on both simulated and real-world datasets demonstrate that EvFocus outperforms existing methods across varying lighting conditions and blur sizes, proving its robustness and practical applicability in event-based defocus imaging.
APA
Zhu, L., Ma, X., Wang, X., Wang, L. & Huang, H.. (2025). EvFocus: Learning to Reconstruct Sharp Images from Out-of-Focus Event Streams. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:79963-79984 Available from https://proceedings.mlr.press/v267/zhu25n.html.

Related Material