Saved in:
Bibliographic Details
Main Authors: Meunier, Etienne, Bouthemy, Patrick
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2310.01040
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910412796067840
author Meunier, Etienne
Bouthemy, Patrick
author_facet Meunier, Etienne
Bouthemy, Patrick
contents Human beings have the ability to continuously analyze a video and immediately extract the motion components. We want to adopt this paradigm to provide a coherent and stable motion segmentation over the video sequence. In this perspective, we propose a novel long-term spatio-temporal model operating in a totally unsupervised way. It takes as input the volume of consecutive optical flow (OF) fields, and delivers a volume of segments of coherent motion over the video. More specifically, we have designed a transformer-based network, where we leverage a mathematically well-founded framework, the Evidence Lower Bound (ELBO), to derive the loss function. The loss function combines a flow reconstruction term involving spatio-temporal parametric motion models combining, in a novel way, polynomial (quadratic) motion models for the spatial dimensions and B-splines for the time dimension of the video sequence, and a regularization term enforcing temporal consistency on the segments. We report experiments on four VOS benchmarks, demonstrating competitive quantitative results, while performing motion segmentation on a whole sequence in one go. We also highlight through visual results the key contributions on temporal consistency brought by our method.
format Preprint
id arxiv_https___arxiv_org_abs_2310_01040
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Segmenting the motion components of a video: A long-term unsupervised model
Meunier, Etienne
Bouthemy, Patrick
Computer Vision and Pattern Recognition
Human beings have the ability to continuously analyze a video and immediately extract the motion components. We want to adopt this paradigm to provide a coherent and stable motion segmentation over the video sequence. In this perspective, we propose a novel long-term spatio-temporal model operating in a totally unsupervised way. It takes as input the volume of consecutive optical flow (OF) fields, and delivers a volume of segments of coherent motion over the video. More specifically, we have designed a transformer-based network, where we leverage a mathematically well-founded framework, the Evidence Lower Bound (ELBO), to derive the loss function. The loss function combines a flow reconstruction term involving spatio-temporal parametric motion models combining, in a novel way, polynomial (quadratic) motion models for the spatial dimensions and B-splines for the time dimension of the video sequence, and a regularization term enforcing temporal consistency on the segments. We report experiments on four VOS benchmarks, demonstrating competitive quantitative results, while performing motion segmentation on a whole sequence in one go. We also highlight through visual results the key contributions on temporal consistency brought by our method.
title Segmenting the motion components of a video: A long-term unsupervised model
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2310.01040