Predicting Deeper into the Future of Semantic Segmentation

Pauline Luc 1, 2 Natalia Neverova 1 Camille Couprie 1 Jakob Verbeek 2 Yann Lecun 3
1 FAIR - Facebook AI Research [Paris]
2 Thoth - Apprentissage de modèles à partir de données massives
LJK - Laboratoire Jean Kuntzmann, Inria Grenoble - Rhône-Alpes
3 New York University [New York]
Abstract : The ability to predict and therefore to anticipate the future is an important attribute of intelligence. It is also of utmost importance in real-time systems, e.g. in robotics or autonomous driving, which depend on visual scene understanding for decision making. While prediction of the raw RGB pixel values in future video frames has been studied in previous work, here we introduce the novel task of predicting semantic segmentations of future frames. Given a sequence of video frames, our goal is to predict segmentation maps of not yet observed video frames that lie up to a second or further in the future. We develop an autoregressive convolutional neural network that learns to iteratively generate multiple frames. Our results on the Cityscapes dataset show that directly predicting future segmentations is substantially better than predicting and then segmenting future RGB frames. Prediction results up to half a second in the future are visually convincing and are much more accurate than those of a baseline based on warping semantic segmentations using optical flow.
Type de document :
Communication dans un congrès
ICCV 2017 - International Conference on Computer Vision, Oct 2017, Venise, Italy. IEEE, pp.648-657, 〈10.1109/ICCV.2017.77〉
Liste complète des métadonnées

HAL-logo.png
HAL-logo.png
17results.jpg
17results.jpg
HAL-logo.jpg
HAL-logo.jpg
https://hal.inria.fr/hal-01494296
Contributeur : Pauline Luc <>
Soumis le : lundi 21 août 2017 - 13:37:02
Dernière modification le : mercredi 20 mars 2019 - 01:29:02

Fichiers

0270.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

LJK | PERSYVAL-LAB | INRIA2017 | INRIA | UGA | LJK-GI-THOTH | LJK_GI

Citation

Pauline Luc, Natalia Neverova, Camille Couprie, Jakob Verbeek, Yann Lecun. Predicting Deeper into the Future of Semantic Segmentation. ICCV 2017 - International Conference on Computer Vision, Oct 2017, Venise, Italy. IEEE, pp.648-657, 〈10.1109/ICCV.2017.77〉. 〈hal-01494296v2〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

1048

Téléchargements de fichiers

2813

Contact


archive-ouverte@inria.fr