TY - GEN
T1 - A comparison of scene flow estimation paradigms
AU - Tsekourakis, Iraklis
AU - Mordohai, Philippos
N1 - Publisher Copyright:
© Springer Nature Switzerland AG 2019.
PY - 2019
Y1 - 2019
N2 - This paper presents a comparison between two core paradigms for computing scene flow from multi-view videos of dynamic scenes. In both approaches, shape and motion estimation are decoupled, in accordance to a large segment of the relevant literature. The first approach is faster and considers only one optical flow field and the depth difference between pixels in consecutive frames to generate a dense scene flow estimate. The second approach is more robust to outliers by considering multiple optical flow fields to generate scene flow. Our goal is to compare the isolated fundamental scene flow estimation methods, without using any post-processing, or optimization. We assess the accuracy of the two methods performing two tests: an optical flow prediction, and a future image prediction, both on a novel view. This is the first quantitative evaluation of scene flow estimation on real imagery of dynamic scenes, in absence of ground truth data.
AB - This paper presents a comparison between two core paradigms for computing scene flow from multi-view videos of dynamic scenes. In both approaches, shape and motion estimation are decoupled, in accordance to a large segment of the relevant literature. The first approach is faster and considers only one optical flow field and the depth difference between pixels in consecutive frames to generate a dense scene flow estimate. The second approach is more robust to outliers by considering multiple optical flow fields to generate scene flow. Our goal is to compare the isolated fundamental scene flow estimation methods, without using any post-processing, or optimization. We assess the accuracy of the two methods performing two tests: an optical flow prediction, and a future image prediction, both on a novel view. This is the first quantitative evaluation of scene flow estimation on real imagery of dynamic scenes, in absence of ground truth data.
UR - http://www.scopus.com/inward/record.url?scp=85065882211&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85065882211&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-19816-9_1
DO - 10.1007/978-3-030-19816-9_1
M3 - Conference contribution
AN - SCOPUS:85065882211
SN - 9783030198152
T3 - Communications in Computer and Information Science
SP - 3
EP - 19
BT - Representations, Analysis and Recognition of Shape and Motion from Imaging Data - 7th International Workshop, RFMI 2017, Revised Selected Papers
A2 - Ghorbel, Faouzi
A2 - Ben Amor, Boulbaba
A2 - Chen, Liming
T2 - 7th International Workshop on Representations, Analysis and Recognition of Shape and Motion from Imaging Data, RFMI 2017
Y2 - 17 December 2017 through 20 December 2017
ER -