Focal Frame Loss: A Simple but Effective Loss for Precipitation Nowcasting

Zhifeng Ma, Hao Zhang, Jie Liu

Abstract

Precipitation nowcasting is an important but hard problem. Currently, with the landing of deep learning, it has been treated as an image prediction problem based on radar echo maps. However, deep learning models suffer from poor performance and blurred prediction results. Lots of improvement works enhance the model by adding complex modules, which increases insufferable training memory and time overhead. Others tempt to add more limitations or guidances on loss, but they usually have little effect in such an extremely complex and difficult task. In this article, we propose a simple but effective loss named focal frame loss (FFL), which assigns different weights to the images in the prediction sequence to focus on the images that are relatively difficult to predict. Experiments on two large-scale radar datasets show that FFL can greatly improve the performance of multiple popular models without introducing additional training costs.
BibTeX
@article{ma2022focal,
  author = {Zhifeng Ma and Hao Zhang and Jie Liu},
  title = {{Focal Frame Loss: A Simple but Effective Loss for Precipitation Nowcasting}},
  journal = {IEEE JSTARS},
  year = {2022},
  doi = {10.1109/jstars.2022.3194522},
}

Loading related content...

Figure