Large Kernel Matters - Improve Semantic Segmentation by Global Convolutional Network

One of recent trends [31, 32, 14] in network architecture design is stacking small filters (e.g., 1×1 or 3×3) in the entire network because the stacked small filters is more efficient than a large kernel, given the same computational complexity. However, in the field of semantic segmentation, where...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) s. 1743 - 1751
Hlavní autoři: Chao Peng, Xiangyu Zhang, Gang Yu, Guiming Luo, Jian Sun
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.07.2017
Témata:
Computational modeling
Feature extraction
Image segmentation
Kernel
Semantics
ISSN:1063-6919, 1063-6919
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:One of recent trends [31, 32, 14] in network architecture design is stacking small filters (e.g., 1×1 or 3×3) in the entire network because the stacked small filters is more efficient than a large kernel, given the same computational complexity. However, in the field of semantic segmentation, where we need to perform dense per-pixel prediction, we find that the large kernel (and effective receptive field) plays an important role when we have to perform the classification and localization tasks simultaneously. Following our design principle, we propose a Global Convolutional Network to address both the classification and localization issues for the semantic segmentation. We also suggest a residual-based boundary refinement to further refine the object boundaries. Our approach achieves state-of-art performance on two public benchmarks and significantly outperforms previous results, 82.2% (vs 80.2%) on PASCAL VOC 2012 dataset and 76.9% (vs 71.8%) on Cityscapes dataset.
ISSN:1063-6919
1063-6919
DOI:10.1109/CVPR.2017.189