Instance Segmentation by Jointly Optimizing Spatial Embeddings and Clustering Bandwidth

Davy Neven, Bert De Brabandere, Marc Proesmans, Luc Van Gool; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 8837-8845

Abstract


Current state-of-the-art instance segmentation methods are not suited for real-time applications like autonomous driving, which require fast execution times at high accuracy. Although the currently dominant proposal-based methods have high accuracy, they are slow and generate masks at a fixed and low resolution. Proposal-free methods, by contrast, can generate masks at high resolution and are often faster, but fail to reach the same accuracy as the proposal-based methods. In this work we propose a new clustering loss function for proposal-free instance segmentation. The loss function pulls the spatial embeddings of pixels belonging to the same instance together and jointly learns an instance-specific clustering bandwidth, maximizing the intersection-over-union of the resulting instance mask. When combined with a fast architecture, the network can perform instance segmentation in real-time while maintaining a high accuracy. We evaluate our method on the challenging Cityscapes benchmark and achieve top results (5% improvement over Mask R-CNN) at more than 10 fps on 2MP images.

Related Material


[pdf] [video]
[bibtex]
@InProceedings{Neven_2019_CVPR,
author = {Neven, Davy and Brabandere, Bert De and Proesmans, Marc and Gool, Luc Van},
title = {Instance Segmentation by Jointly Optimizing Spatial Embeddings and Clustering Bandwidth},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}