Off-the-shelf sensor integration for mono-SLAM on smart devices

Tiefenbacher, Philipp; Schulze, Timo; Rigoll, Gerhard

doi:10.1109/cvprw.2015.7301361

Cited by 7 publications

(8 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their unscented Kalman filter (UKF) considered gyroscope and accelerometer measurements. Tiefenbacher et al [1] used the IMU data as control input for the UKF. Furthermore, a motion model based on the a priori estimate of the UKF was presented.…”

Section: Related Workmentioning

confidence: 99%

“…The precision and stability of the motion model has a direct influence on tracking performance. For our experiments, we make use of the PTAM motion model (PMM) and the findings in [1].…”

Section: Sensor Fusionmentioning

confidence: 99%

“…This motion model is used for both PTAM and the PTAM+ [1] variants. For the latter, an unscented Kalman filter (UKF) additionally updates the final PTAM pose leading to a posterior pose estimate (similar to [1]). The state vector is identical to [1].…”

Section: Sensor Fusionmentioning

confidence: 99%

“…For the latter, an unscented Kalman filter (UKF) additionally updates the final PTAM pose leading to a posterior pose estimate (similar to [1]). The state vector is identical to [1]. It has 26 dimensions and contains, among others, the IMU-to-world attitude, the position and velocity vectors as well as the gravity vector.…”

Section: Sensor Fusionmentioning

confidence: 99%

“…Handhelds are ubiquitous and are usually equipped with a video camera which enables the integration of simultaneous localization and mapping (SLAM). Handhelds also include additional sensors, the inertial measurement units (IMUs), which can improve the SLAM accuracy [1].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Sensor Fusion for Sparse SLAM with Descriptor Pooling

Tiefenbacher

Heuser

Schulze

et al. 2016

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

This paper focuses on the advancement of a monocular sparse-SLAM algorithm via two techniques: Local feature maintenance and descriptor-based sensor fusion. We present two techniques that maintain the descriptor of a local feature: Pooling and bestfit. The maintenance procedure aims at defining more accurate descriptors, increasing matching performance and thereby tracking accuracy. Moreover, sensors besides the camera can be used to improve tracking robustness and accuracy via sensor fusion. State-of-the-art sensor fusion techniques can be divided into two categories. They either use a Kalman filter that includes sensor data in its state vector to conduct a posterior pose update, or they create world-aligned image descriptors with the help of the gyroscope. This paper is the first to compare and combine these two approaches. We release a new evaluation dataset which comprises 21 scenes that include a dense ground truth trajectory, IMU data, and camera data. The results indicate that descriptor pooling significantly improves pose accuracy. Furthermore, we show that descriptor-based sensor fusion outperforms Kalman filter-based approaches (EKF and UKF).

show abstract

Section: Related Workmentioning

confidence: 99%

“…The precision and stability of the motion model has a direct influence on tracking performance. For our experiments, we make use of the PTAM motion model (PMM) and the findings in [1].…”

Section: Sensor Fusionmentioning

confidence: 99%