Skip to Main content Skip to Navigation
Journal articles

WatchPose: A View-Aware Approach for Camera Pose Data Collection in Industrial Environments

Cong Yang Gilles Simon 1, 2 John See Marie-Odile Berger 1 Wenyong Wang
1 MAGRIT - Visual Augmentation of Complex Environments
Inria Nancy - Grand Est, LORIA - ALGO - Department of Algorithms, Computation, Image and Geometry
2 MAGRIT-POST - Augmentation visuelle d'environnements complexes
Inria Nancy - Grand Est, LORIA - ALGO - Department of Algorithms, Computation, Image and Geometry
Abstract : Collecting correlated scene images and camera poses is an essential step towards learning absolute camera pose regression models. While the acquisition of such data in living environments is relatively easy by following regular roads and paths, it is still a challenging task in constricted industrial environments. This is because industrial objects have varied sizes and inspections are usually carried out with non-constant motions. As a result, regression models are more sensitive to scene images with respect to viewpoints and distances. Motivated by this, we present a simple but efficient camera pose data collection method, WatchPose, to improve the generalization and robustness of camera pose regression models. Specifically, WatchPose tracks nested markers and visualizes viewpoints in an Augmented Reality-(AR) based manner to properly guide users to collect training data from broader camera-object distances and more diverse views around the objects. Experiments show that WatchPose can effectively improve the accuracy of existing camera pose regression models compared to the traditional data acquisition method. We also introduce a new dataset, Industrial10, to encourage the community to adapt camera pose regression methods for more complex environments.
Complete list of metadata

Cited literature [54 references]  Display  Hide  Download

https://hal.inria.fr/hal-02735272
Contributor : Gilles Simon <>
Submitted on : Tuesday, June 2, 2020 - 3:14:52 PM
Last modification on : Tuesday, May 18, 2021 - 3:44:10 PM
Long-term archiving on: : Wednesday, December 2, 2020 - 2:54:43 PM

File

sensors-20-03045.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Cong Yang, Gilles Simon, John See, Marie-Odile Berger, Wenyong Wang. WatchPose: A View-Aware Approach for Camera Pose Data Collection in Industrial Environments. Sensors, MDPI, 2020, 20 (11), ⟨10.3390/s20113045⟩. ⟨hal-02735272⟩

Share

Metrics

Record views

83

Files downloads

311