Naslov
Towards Keypoint Guided Self-Supervised Depth Estimation

Autori
Bartol, Kristijan ; Bojanić, David ; Petković, Tomislav ; Pribanić, Tomislav ; Donoso, Yago

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - (Volume 4) / Farinella, Giovanni Maria ; Radeva, Petia ; Braz, Jose - : SCITEPRESS, 2020, 583-589

ISBN
978-989-758-402-2

Skup
15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020)

Mjesto i datum
Valletta, Malta, 27.02.2020. - 29.02.2020

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Monocular Depth Estimation ; Self-supervised Learning ; Keypoint Similarity Loss

Sažetak
This paper proposes to use keypoints as a self- supervision clue for learning depth map estimation from a collection of input images. As ground truth depth from real images is difficult to obtain, there are many unsupervised and self- supervised approaches to depth estimation that have been proposed. Most of these unsupervised approaches use depth map and ego-motion estimations to reproject the pixels from the current image into the adjacent image from the image collection. Depth and ego-motion estimations are evaluated based on pixel intensity differences between the correspondent original and reprojected pixels. Instead of reprojecting the individual pixels, we propose to first select image keypoints in both images and then reproject and compare the correspondent keypoints of the two images. The keypoints should describe the distinctive image features well. By learning a deep model with and without the keypoint extraction technique, we show that using the keypoints improve the depth e stimation learning. We also propose some future directions for keypoint-guided learning of structure-from- motion problems.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo

POVEZANOST RADA