Method Details

Details for method 'Axial-DeepLab-L [Cityscapes-fine]'


Method overview

name Axial-DeepLab-L [Cityscapes-fine]
challenge pixel-level semantic labeling
details Convolution exploits locality for efficiency at a cost of missing long range context. Self-attention has been adopted to augment CNNs with non-local interactions. Recent works prove it possible to stack self-attention layers to obtain a fully attentional network by restricting the attention to a local region. In this paper, we attempt to remove this constraint by factorizing 2D self-attention into two 1D self-attentions. This reduces computation complexity and allows performing attention within a larger or even global region. In companion, we also propose a position-sensitive self-attention design. Combining both yields our position-sensitive axial-attention layer, a novel building block that one could stack to form axial-attention models for image classification and dense prediction. We demonstrate the effectiveness of our model on four large-scale datasets. In particular, our model outperforms all existing stand-alone self-attention models on ImageNet. Our Axial-DeepLab improves 2.8% PQ over bottom-up state-of-the-art on COCO test-dev. This previous state-of-the-art is attained by our small variant that is 3.8x parameter-efficient and 27x computation-efficient. Axial-DeepLab also achieves state-of-the-art results on Mapillary Vistas and Cityscapes.
publication Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation
Huiyu Wang, Yukun Zhu, Bradley Green, Hartwig Adam, Alan Yuille, Liang-Chieh Chen
ECCV 2020 (spotlight)
project page / code
used Cityscapes data fine annotations
used external data ImageNet
runtime n/a
subsampling no
submission date March, 2020
previous submissions


Average results

Metric Value
IoU Classes 79.4968
iIoU Classes 57.4839
IoU Categories 91.4892
iIoU Categories 76.2137


Class results

Class IoU iIoU
road 98.6313 -
sidewalk 86.5022 -
building 93.4365 -
wall 52.0461 -
fence 61.3156 -
pole 70.1632 -
traffic light 77.5351 -
traffic sign 81.0828 -
vegetation 93.4519 -
terrain 72.3189 -
sky 95.6783 -
person 87.9289 68.2537
rider 75.6083 55.4641
car 96.0295 84.76
truck 68.4795 39.7625
bus 81.3838 55.8922
train 77.0785 51.1847
motorcycle 71.0344 51.0277
bicycle 70.7347 53.5266


Category results

Category IoU iIoU
flat 98.6638 -
nature 93.1728 -
object 76.0034 -
sky 95.6783 -
construction 93.6072 -
human 88.0594 69.6875
vehicle 95.2395 82.7399



Download results as .csv file

Benchmark page