arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.04812cs.CV
收藏

基于视觉词汇学习和混合池的弱监督语义分割

Weakly-Supervised Semantic Segmentation with Visual Words Learning and Hybrid Pooling

Wuhan University(武汉大学);JD Explore Academy, JD.com(京东探索研究院)

Lixiang Ru, Bo Du, Yibing Zhan, Chen Wu

摘要:

具有图像级标签的弱监督语义分割(WSSS)方法通常训练分类网络以生成类激活图(CAM)作为初始粗分割标签。然而,目前的WSSS方法仍然远远不能令人满意地执行,因为它们所采用的CAM 1)通常集中在部分可区分的对象区域和2)通常包含无用的背景区域。这两个问题归因于训练分类网络时的唯一图像级监督和全局信息的聚合。在这项工作中,我们提出了视觉词汇学习模块和混合池化方法,并将它们纳入分类网络中,以缓解上述问题。在视觉词学习模块中,我们通过强制分类网络学习细粒度的视觉词标签来解决第一个问题,以便可以发现更多的对象范围。具体地说,视觉词汇学习与码本,它可以通过两个建议的策略,即基于学习的策略和记忆库策略更新。CAM的第二个缺点是缓解与建议的混合池,它结合了全球平均值和本地判别信息,同时确保对象的完整性和减少背景区域。我们在PASCAL VOC 2012和MS COCO 2014数据集上评估了我们的方法。在没有任何额外显著性的情况下,我们的方法在PASCAL VOC数据集的$val$和$test$集上分别实现了70.6%和70.7%的mIoU,在MS COCO数据集的$val$集上实现了36.2%的mIoU,这显著超过了最先进的WSSS方法的性能。

英文摘要:

Weakly-Supervised Semantic Segmentation (WSSS) methods with image-level labels generally train a classification network to generate the Class Activation Maps (CAMs) as the initial coarse segmentation labels. However, current WSSS methods still perform far from satisfactorily because their adopted CAMs 1) typically focus on partial discriminative object regions and 2) usually contain useless background regions. These two problems are attributed to the sole image-level supervision and aggregation of global information when training the classification networks. In this work, we propose the visual words learning module and hybrid pooling approach, and incorporate them in the classification network to mitigate the above problems. In the visual words learning module, we counter the first problem by enforcing the classification network to learn fine-grained visual word labels so that more object extents could be discovered. Specifically, the visual words are learned with a codebook, which could be updated via two proposed strategies, i.e. learning-based strategy and memory-bank strategy. The second drawback of CAMs is alleviated with the proposed hybrid pooling, which incorporates the global average and local discriminative information to simultaneously ensure object completeness and reduce background regions. We evaluated our methods on PASCAL VOC 2012 and MS COCO 2014 datasets. Without any extra saliency prior, our method achieved 70.6% and 70.7% mIoU on the $val$ and $test$ set of PASCAL VOC dataset, respectively, and 36.2% mIoU on the $val$ set of MS COCO dataset, which significantly surpassed the performance of state-of-the-art WSSS methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑