An End to End Learning Approach for Distance Estimation Trained with Artificially Generated Stereo Images
An End to End Learning Approach for Distance Estimation Trained with Artificially Generated Stereo Images
Autori:
Izdanje: International Scientific Conference on Information Technology and Data Related Research
DOI: 10.15308/Sinteza-2020-3-7
Oblast: Artificial Intelligence Atlas
Stranice: 3-7
Apstrakt:
This paper proposes a solution for distance estimation using stereo images. The solution is a convolutional neural network that takes two images as an input, and outputs the distance estimate, without the need for prior camera calibration or disparity map calculation. The dataset used for training consists of images generated from an artificially constructed 3D scene. The training algorithm used was stochastic gradient descent. Evaluation of the solution was conducted on a separate dataset. Mean absolute error after the evaluation was 1.59 m, while the median value of the absolute error was 1.2 m. These results show that the proposed solution is a valid proof of concept for the usage of convolutional neural networks for the distance estimation of objects in stereo images in a single step.
Ključne reči: artificially generated data, convolutional neural networks, stereo vision
Priložene datoteke:
- 3-7 ( veličina: 523,55 KB, broj pregleda: 354 )
Zahvaljujemo se što ste preuzeli publikaciju sa portala Singipedia.
Ukoliko želite da se prijavite za obaveštenja o sadržajima iz oblasti ove publikacije, možete nam ostaviti adresu svoje elektronske pošte.
Preuzimanje citata:
BibTeX format
RefWorks Tagged format
Unapred formatirani prikaz citata
BibTeX format
@article{article, author = {N. Nešić, M. Vidović, I. Radosavljević, A. Mitrović and �. Obradović}, title = {An End to End Learning Approach for Distance Estimation Trained with Artificially Generated Stereo Images}, journal = {International Scientific Conference on Information Technology and Data Related Research}, year = 2020, pages = {3-7}, doi = {10.15308/Sinteza-2020-3-7} }
RT Conference Proceedings A1 Nebojša Nešić A1 Mladen Vidović A1 Ivan Radosavljević A1 Aleksandra Mitrović A1 Đorđe Obradović T1 An End to End Learning Approach for Distance Estimation Trained with Artificially Generated Stereo Images AD Univerzitet Singidunum, Beograd, Beograd, Srbija YR 2020 NO doi: 10.15308/Sinteza-2020-3-7
N. Nešić, M. Vidović, I. Radosavljević, A. Mitrović and . Obradović, An End to End Learning Approach for Distance Estimation Trained with Artificially Generated Stereo Images, Univerzitet Singidunum, Beograd, 2020, doi:10.15308/Sinteza-2020-3-7