Method Details

Details for method 'Hierarchical Multi-Scale Attention for Semantic Segmentation'


Method overview

name Hierarchical Multi-Scale Attention for Semantic Segmentation
challenge pixel-level semantic labeling
details Multi-scale inference is commonly used to improve the results of semantic segmentation. Multiple images scales are passed through a network and then the results are combined with averaging or max pooling. In this work, we present an attention-based approach to combining multi-scale predictions. We show that predictions at certain scales are better at resolving particular failures modes and that the network learns to favor those scales for such cases in order to generate better predictions. Our attention mechanism is hierarchical, which enables it to be roughly 4x more memory efficient to train than other recent approaches. In addition to enabling faster training, this allows us to train with larger crop sizes which leads to greater model accuracy. We demonstrate the result of our method on two datasets: Cityscapes and Mapillary Vistas. For Cityscapes, which has a large number of weakly labelled images, we also leverage auto-labelling to improve generalization. Using our approach we achieve a new state-of-the-art results in both Mapillary (61.1 IOU val) and Cityscapes (85.1 IOU test).
publication Hierarchical Multi-Scale Attention for Semantic Segmentation
Andrew Tao, Karan Sapra, Bryan Catanzaro
project page / code
used Cityscapes data fine annotations, coarse annotations
used external data Mapillary
runtime n/a
subsampling no
submission date February, 2020
previous submissions


Average results

Metric Value
IoU Classes 85.1335
iIoU Classes 69.9886
IoU Categories 93.1479
iIoU Categories 85.4237


Class results

Class IoU iIoU
road 98.9807 -
sidewalk 89.2472 -
building 94.9115 -
wall 71.627 -
fence 69.1209 -
pole 75.8215 -
traffic light 82.0459 -
traffic sign 85.24 -
vegetation 94.4928 -
terrain 74.9946 -
sky 96.3307 -
person 90.0386 78.5226
rider 79.3819 63.1778
car 96.9262 92.7017
truck 79.8308 56.5522
bus 93.9764 69.6212
train 85.7651 64.0298
motorcycle 77.4489 62.8654
bicycle 81.3566 72.438


Category results

Category IoU iIoU
flat 98.9103 -
nature 94.2971 -
object 80.827 -
sky 96.3307 -
construction 94.9303 -
human 90.114 79.3571
vehicle 96.6257 91.4902



Download results as .csv file

Benchmark page