In semi-supervised semantic segmentation, the Mean Teacher- and
co-training-based approaches are employed to mitigate confirmation bias and
coupling problems. However, despite their high performance, these approaches
frequently involve complex training pipelines and a substantial computational
burden, limiting the scalability and compatibility of these methods. In this
paper, we propose a PrevMatch framework that effectively mitigates the
aforementioned limitations by maximizing the utilization of the temporal
knowledge obtained during the training process. The PrevMatch framework relies
on two core strategies: (1) we reconsider the use of temporal knowledge and
thus directly utilize previous models obtained during training to generate
additional pseudo-label guidance, referred to as previous guidance. (2) we
design a highly randomized ensemble strategy to maximize the effectiveness of
the previous guidance. Experimental results on four benchmark semantic
segmentation datasets confirm that the proposed method consistently outperforms
existing methods across various evaluation protocols. In particular, with
DeepLabV3+ and ResNet-101 network settings, PrevMatch outperforms the existing
state-of-the-art method, Diverse Co-training, by +1.6 mIoU on Pascal VOC with
only 92 annotated images, while achieving 2.4 times faster training.
Furthermore, the results indicate that PrevMatch induces stable optimization,
particularly in benefiting classes that exhibit poor performance. Code is
available at this https URL

提出了一种 PrevMatch 框架，通过最大化利用训练过程中获得的时间知识，有效地缓解了复杂的训练流程和计算负担所带来的限制，同时在各种评估协议上始终优于现有方法，特别是在 DeepLabV3 + 和 ResNet-101 网络设置下，在只有 92 个标注图像的 Pascal VOC 数据集上，与现有最先进方法 Diverse Co-training 相比，mIoU 提升了 1.6，同时训练速度增快了 2.4 倍，并且结果表明 PrevMatch 能够稳定地优化，特别适用于性能较差的类别。