Abstract | ||
---|---|---|
Abstract Designed as extremely deep architectures, deep residual networks which provide a rich visual representation and offer robust convergence behaviors have recently achieved exceptional performance in numerous computer vision problems. Being directly applied to a scene labeling problem, however, they were limited to capture long-range contextual dependence, which is a critical aspect. To address this issue, we propose a novel approach, Contextual Recurrent Residual Networks (CRRN) which is able to simultaneously handle rich visual representation learning and long-range context modeling within a fully end-to-end deep network. Furthermore, our proposed end-to-end CRRN is completely trained from scratch, without using any pre-trained models in contrast to most existing methods usually fine-tuned from the state-of-the-art pre-trained models, e.g. VGG-16, ResNet, etc. The experiments are conducted on four challenging scene labeling datasets, i.e. SiftFlow, CamVid, Stanford background and SUN datasets, and compared against various state-of-the-art scene labeling methods. |
Year | Venue | DocType |
---|---|---|
2018 | Pattern Recognition | Journal |
Volume | Citations | PageRank |
abs/1704.03594 | 5 | 0.41 |
References | Authors | |
36 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
T. Hoang Ngan Le | 1 | 178 | 14.25 |
Chi Nhan Duong | 2 | 37 | 10.68 |
Ligong Han | 3 | 5 | 2.44 |
Khoa Luu | 4 | 200 | 26.05 |
Marios Savvides | 5 | 1485 | 112.94 |
Dipan K. Pal | 6 | 76 | 7.71 |