Name
Abstract | ||
|---|---|---|
Most approaches for instance-aware semantic labeling traditionally focus on accuracy. Other aspects like runtime and memory footprint are arguably as important for real-time applications such as autonomous driving. Motivated by this observation and inspired by recent works that tackle multiple tasks with a single integrated architecture, in this paper we present a real-time efficient implementation based on ENet that solves three autonomous driving related tasks at once: semantic scene segmentation, instance segmentation and monocular depth estimation. Our approach builds upon a branched ENet architecture with a shared encoder but different decoder branches for each of the three tasks. The presented method can run at 21 fps at a resolution of 1024x512 on the Cityscapes dataset without sacrificing accuracy compared to running each task separately. |
Year | Venue | Field |
|---|---|---|
2017 | arXiv: Computer Vision and Pattern Recognition | Architecture,Computer science,Segmentation,Semantic labeling,Integrated architecture,Artificial intelligence,Encoder,Memory footprint,Monocular,Scene segmentation,Machine learning |
DocType | Volume | Citations |
Journal | abs/1708.02550 | 6 |
PageRank | References | Authors |
0.46 | 9 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Davy Neven | 1 | 28 | 1.15 |
| Bert De Brabandere | 2 | 28 | 1.15 |
| Stamatios Georgoulis | 3 | 109 | 10.21 |
| Marc Proesmans | 4 | 277 | 34.37 |
| Luc Van Gool | 5 | 27566 | 1819.51 |