EvFlow-GS: Event Enhanced Motion Deblurring with Optical Flow for 3D Gaussian Splatting
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913058926886912 |
|---|---|
| author | An, Feiyu Deng, Yufei Zhang, Zihui Xiao, Rong |
| author_facet | An, Feiyu Deng, Yufei Zhang, Zihui Xiao, Rong |
| contents | Achieving sharp 3D reconstruction from motion-blurred images alone becomes challenging, motivating recent methods to incorporate event cameras, benefiting from microsecond temporal resolution. However, they suffer from residual artifacts and blurry texture details due to misleading supervision from inaccurate event double integral priors and noisy, blurry events. In this study, we propose EvFlow-GS, a unified framework that leverages event streams and optical flow to optimize an end-to-end learnable double integral (LDI), camera poses, and 3D Gaussian Splatting (3DGS) jointly on-the-fly. Specifically, we first extract edge information from the events using optical flow and then formulate a novel event-based loss applied separately to different modules. Additionally, we exploit a novel event-residual prior to strengthen the supervision of intensity changes between images rendered from 3DGS. Finally, we integrate the outputs of both 3DGS and LDI into a joint loss, enabling their optimization to mutually facilitate each other. Experiments demonstrate the leading performance of our EvFlow-GS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_22183 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EvFlow-GS: Event Enhanced Motion Deblurring with Optical Flow for 3D Gaussian Splatting An, Feiyu Deng, Yufei Zhang, Zihui Xiao, Rong Computer Vision and Pattern Recognition Achieving sharp 3D reconstruction from motion-blurred images alone becomes challenging, motivating recent methods to incorporate event cameras, benefiting from microsecond temporal resolution. However, they suffer from residual artifacts and blurry texture details due to misleading supervision from inaccurate event double integral priors and noisy, blurry events. In this study, we propose EvFlow-GS, a unified framework that leverages event streams and optical flow to optimize an end-to-end learnable double integral (LDI), camera poses, and 3D Gaussian Splatting (3DGS) jointly on-the-fly. Specifically, we first extract edge information from the events using optical flow and then formulate a novel event-based loss applied separately to different modules. Additionally, we exploit a novel event-residual prior to strengthen the supervision of intensity changes between images rendered from 3DGS. Finally, we integrate the outputs of both 3DGS and LDI into a joint loss, enabling their optimization to mutually facilitate each other. Experiments demonstrate the leading performance of our EvFlow-GS. |
| title | EvFlow-GS: Event Enhanced Motion Deblurring with Optical Flow for 3D Gaussian Splatting |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.22183 |