Unabridged adjacent modulation for clothing parsing

作者：

Highlights：

• Based on the encoder-decoder architecture, we propose a block-specific unabridged channel attention mechanism, such that features within each block can be recalibrated.

• A top-down adjacent modulation for decoder network is proposed so that the lowlevel features substantially contain abundant semantic contexts.

• We demonstrate the effectiveness of UAM-Net via two challenging benchmarks. Results declare that we can achieve a new state-of-the-art on colorful fashion parsing dataset and comparable performance on modified fashion clothing dataset with less computation overhead.

摘要

•Based on the encoder-decoder architecture, we propose a block-specific unabridged channel attention mechanism, such that features within each block can be recalibrated.•A top-down adjacent modulation for decoder network is proposed so that the lowlevel features substantially contain abundant semantic contexts.•We demonstrate the effectiveness of UAM-Net via two challenging benchmarks. Results declare that we can achieve a new state-of-the-art on colorful fashion parsing dataset and comparable performance on modified fashion clothing dataset with less computation overhead.

论文关键词：Encoder-decoder network,Clothing parsing,Attention learning,Features modulation,Self-supervised learning

论文评审过程：Received 15 March 2021, Revised 16 January 2022, Accepted 16 February 2022, Available online 24 February 2022, Version of Record 10 March 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108594