论文标题
ASYINST：与Depthgrad的不对称亲和力和盒子的颜色，用于盒子监督实例分段
AsyInst: Asymmetric Affinity with DepthGrad and Color for Box-Supervised Instance Segmentation
论文作者
论文摘要
弱监督的实例细分是一项具有挑战性的任务。现有方法通常使用边界框作为监督，并使用正规化损失项（例如成对颜色亲和力损失）进行优化网络。通过系统的分析，我们发现常用的成对亲和力损失具有两个局限性：（1）它与颜色亲和力一起使用，但导致与其他方式（例如深度梯度）的性能较低，（2）原始亲和力损失并不能阻止预期的琐碎预测，但实际上由于亲和力损失术语是对称的，因此实际上会加速此过程。为了克服这两个局限性，在本文中，我们提出了一种新型的不对称亲和力损失，该损失对微不足道的预测提供了惩罚，并以不同方式的亲和力损失很好地概括了。通过提出的不对称亲和力损失，我们的方法优于CityScapes数据集上的最先进方法，并在Mask AP中优于基线方法3.5％。
The weakly supervised instance segmentation is a challenging task. The existing methods typically use bounding boxes as supervision and optimize the network with a regularization loss term such as pairwise color affinity loss for instance segmentation. Through systematic analysis, we found that the commonly used pairwise affinity loss has two limitations: (1) it works with color affinity but leads to inferior performance with other modalities such as depth gradient, (2)the original affinity loss does not prevent trivial predictions as intended but actually accelerates this process due to the affinity loss term being symmetric. To overcome these two limitations, in this paper, we propose a novel asymmetric affinity loss which provides the penalty against the trivial prediction and generalizes well with affinity loss from different modalities. With the proposed asymmetric affinity loss, our method outperforms the state-of-the-art methods on the Cityscapes dataset and outperforms our baseline method by 3.5% in mask AP.
