A CNN-based multi-task framework for weather recognition with multi-scale weather cues

作者：

Highlights：

• We capture multi-scale weather-cue features for multi-task weather recognition.

• A generalized pooling is proposed to generate weather feature representations.

• An adaptive weighting scheme is proposed for multi-task weather recognition.

• The state-of-the-art results are reported on two weather recognition datasets.

摘要

•We capture multi-scale weather-cue features for multi-task weather recognition.•A generalized pooling is proposed to generate weather feature representations.•An adaptive weighting scheme is proposed for multi-task weather recognition.•The state-of-the-art results are reported on two weather recognition datasets.

论文关键词：Multi-task weather recognition,Multi-scale weather-cue features,Weather-cue segmentation,Weather classification

论文评审过程：Received 25 May 2021, Revised 20 December 2021, Accepted 17 February 2022, Available online 5 March 2022, Version of Record 10 March 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116689