Predicting Future Instance Segmentation by Forecasting Convolutional Features

Abstract : Anticipating future events is an important prerequisite towards intelligent behavior. Video forecasting has been studied as a proxy task towards this goal. Recent work has shown that to predict semantic segmentation of future frames, forecasting at the semantic level is more effective than forecasting RGB frames and then segmenting these. In this paper we consider the more challenging problem of future instance segmentation, which additionally segments out individual objects. To deal with a varying number of output labels per image, we develop a predictive model in the space of fixed-sized convolutional features of the Mask R-CNN instance segmentation model. We apply the "detection head" of Mask R-CNN on the predicted features to produce the instance segmentation of future frames. Experiments show that this approach significantly improves over strong baselines based on optical flow and repurposed instance segmentation architectures.
Liste complète des métadonnées

https://hal.inria.fr/hal-01757669
Contributor : Pauline Luc <>
Submitted on : Wednesday, October 3, 2018 - 12:19:52 PM
Last modification on : Wednesday, March 20, 2019 - 1:29:04 AM

File

luc18eccv_arxiv.pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Pauline Luc, Camille Couprie, Yann Lecun, Jakob Verbeek. Predicting Future Instance Segmentation by Forecasting Convolutional Features. ECCV 2018 - European Conference on Computer Vision, Sep 2018, Munich, Germany. pp.593-608, ⟨10.1007/978-3-030-01240-3_36⟩. ⟨hal-01757669v2⟩

Share

Metrics

Record views

349

Files downloads

598