Method Details
Details for method 'Axial-DeepLab-XL [Cityscapes-fine]'
Method overview
| name | Axial-DeepLab-XL [Cityscapes-fine] |
| challenge | instance-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) https://arxiv.org/abs/2003.07853 |
| project page / code | https://github.com/csrhddlam/axial-deeplab |
| used Cityscapes data | fine annotations |
| used external data | ImageNet |
| runtime | n/a |
| subsampling | no |
| submission date | March, 2020 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| AP | 33.9951 |
| AP50% | 55.8674 |
| AP100m | 49.6014 |
| AP50m | 53.1034 |
Class results
| Class | AP | AP50% | AP100m | AP50m |
|---|---|---|---|---|
| person | 32.3156 | 61.2127 | 51.4414 | 51.5652 |
| rider | 28.1706 | 59.2245 | 42.4664 | 42.8731 |
| car | 52.6413 | 74.9595 | 74.4193 | 77.103 |
| truck | 32.64 | 41.7007 | 47.502 | 57.4945 |
| bus | 41.7751 | 52.9174 | 63.5619 | 73.4058 |
| train | 38.0636 | 57.9667 | 50.1989 | 54.0729 |
| motorcycle | 25.8692 | 51.9398 | 35.5985 | 36.5713 |
| bicycle | 20.4851 | 47.0178 | 31.6226 | 31.7412 |