VIO-Aided Structure from Motion Under Challenging Environments

Jiang, Zijie; Taira, Hajime; Miyashita, Naoyuki; Okutomi, Masatoshi

doi:10.1109/icit46573.2021.9453630

Cited by 4 publications

(3 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Locher et al [38] filtered wrong epipolar geometries by RA before applying the divideand-conquer method [18]. Jiang et al [39] use a visualinertial navigation system (VINS) [40] to first estimate the camera trajectories with loop detection and loop closure [41]. Images are then divided into sequences according to timestamps.…”

Section: Related Workmentioning

confidence: 99%

“…Images are then divided into sequences according to timestamps. However, [39] requires two carefully designed systems: one for VINS with loop detection and the other for SfM. Loop detection is also a challenge in real-world scenes.…”

Section: Related Workmentioning

confidence: 99%

“…We assume cameras are pre-calibrated and use the ceres-solver [49] for bundle adjustment. We did not compare our method against [39], as VINs [40] fails to find the right loops in our datasets. All methods are run on the same computer with 40 CPU cores and 96 GB RAM.…”

Section: A Implementation Detailsmentioning

confidence: 99%

See 2 more Smart Citations

AdaSfM: From Coarse Global to Fine Incremental Adaptive Structure from Motion

Chen¹,

Yu²,

Song³

et al. 2023

Preprint

View full text Add to dashboard Cite

Despite the impressive results achieved by many existing Structure from Motion (SfM) approaches, there is still a need to improve the robustness, accuracy, and efficiency on large-scale scenes with many outlier matches and sparse view graphs. In this paper, we propose AdaSfM: a coarseto-fine adaptive SfM approach that is scalable to large-scale and challenging datasets. Our approach first does a coarse global SfM which improves the reliability of the view graph by leveraging measurements from low-cost sensors such as Inertial Measurement Units (IMUs) and wheel encoders. Subsequently, the view graph is divided into sub-scenes that are refined in parallel by a fine local incremental SfM regularised by the result from the coarse global SfM to improve the camera registration accuracy and alleviate scene drifts. Finally, our approach uses a threshold-adaptive strategy to align all local reconstructions to the coordinate frame of global SfM. Extensive experiments on large-scale benchmark datasets show that our approach achieves state-of-the-art accuracy and efficiency.

show abstract