|
Article information
2024 , Volume 29, ¹ 2, p.79-94
Letenkov M.A., Cherskikh E.O.
ML based approach for determining the spatial position and size of objects on images
This paper considers an urgent problem of assessing spatial position and geometric characteristics of environmental objects from images. An approach was developed based on combining the results of object detection using the Mask R-CNN model and the reconstruction of depth maps obtained using the RealSense camera. We evaluated the class-averaged values of the relative error in determining the size of objects for test sets of images formed at various levels of scene illumination: 0.1449, 0.3313, 0.6332. Also, within the experiments relative deviation values were obtained when determining the spatial positions of objects: 0.1010, 0.1624, 0.3477.
Keywords: objects size evaluation, spatial position assessing, object detection, depth map reconstruction, Mask R-CNN, Intel RealSense
doi: 10.25743/ICT.2024.29.2.007
Author(s):
Letenkov Maksim Andreevich
Position: Junior Research Scientist
Office: St. Petersburg Federal Research Center of the Russian Academy of Sciences , St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences
Address: 199178, Russia, St-Petersburg, 39, 14th Line, VI
E-mail: letenkovmaksim@yandex.ru
Cherskikh Ekaterina Olegovna
Position: Junior Research Scientist
Office: St. Petersburg Federal Research Center of the Russian Academy of Sciences , St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences
Address: 199178, Russia, St-Petersburg, 39, 14th Line, VI
E-mail: cherskikh.e@iias.spb.su
Bibliography link:
Letenkov M.A., Cherskikh E.O. ML based approach for determining the spatial position and size of objects on images // Computational technologies. 2024. V. 29. ¹ 2. P. 79-94
|
