Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Unblur-SLAM: Dense Neural SLAM for Blurry Inputs

About

We propose Unblur-SLAM, a novel RGB SLAM pipeline for sharp 3D reconstruction from blurred image inputs. In contrast to previous work, our approach is able to handle different types of blur and demonstrates state-of-the-art performance in the presence of both motion blur and defocus blur. Moreover, we adjust the computation effort with the amount of blur in the input image. As a first stage, our method uses a feed-forward image deblurring model for which we propose a suitable training scheme that can improve both tracking and mapping modules. Frames that are successfully deblurred by the feed-forward network obtain refined poses and depth through local-global multi-view optimization and loop closure. Frames that fail the first stage deblurring are directly modeled through the global 3DGS representation and an additional blur network to model multiple blurred sub-frames and simulate the blur formation process in 3D space, thereby learning sharp details and refined sub-frame poses. Experiments on several real-world datasets demonstrate consistent improvements in both pose estimation and sharp reconstruction results of geometry and texture.

Qi Zhang, Denis Rozumny, Francesco Girlanda, Sezer Karaoglu, Marc Pollefeys, Theo Gevers, Martin R. Oswald• 2026

Related benchmarks

TaskDatasetResultRank
3D Scene DeblurringDeblur-NeRF motion blur and defocus blur
PSNR29.49
8
3D Scene DeblurringDeblur-NeRF defocus blur
PSNR27.45
6
Trajectory EstimationTUM
ATE RMSE0.336
4
Trajectory EstimationMCD
ATE RMSE0.128
4
SLAMTUM RGBD (fr1_desk, fr2_xyz, fr3_office)
FPS0.85
4
SLAMArchViz-2
ATE0.0027
2
SLAMArchViz 3
ATE0.0067
2
SLAMArchViz
Average Translational Error0.0056
2
SLAMArchViz 1
ATE0.0075
2
Showing 9 of 9 rows

Other info

Follow for update