Method Details

Details for method 'Axial-DeepLab-L [Mapillary Vistas]'


Method overview

name Axial-DeepLab-L [Mapillary Vistas]
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, Mapillary Vistas
runtime n/a
subsampling no
submission date March, 2020
previous submissions


Average results

Metric Value
IoU Classes 83.1423
iIoU Classes 64.023
IoU Categories 92.3349
iIoU Categories 78.9062


Class results

Class IoU iIoU
road 98.8119 -
sidewalk 87.8007 -
building 94.1522 -
wall 59.7815 -
fence 68.0794 -
pole 73.4194 -
traffic light 79.4888 -
traffic sign 82.6183 -
vegetation 93.948 -
terrain 72.7257 -
sky 96.0848 -
person 88.9012 71.1657
rider 77.5147 58.2308
car 96.4877 87.0429
truck 76.8575 51.2779
bus 91.2153 64.0995
train 92.6248 63.8267
motorcycle 74.7286 57.8978
bicycle 74.4632 58.6424


Category results

Category IoU iIoU
flat 98.7371 -
nature 93.6701 -
object 78.642 -
sky 96.0848 -
construction 94.4179 -
human 88.8617 72.3049
vehicle 95.931 85.5074



Download results as .csv file

Benchmark page