Recent breakthroughs in semi-supervised semantic segmentation have been developed through contrastive learning. In prevalent pixel-wise contrastive learning solutions, the model maps pixels to deterministic representations and regularizes them in the latent space. However, there exist inaccurate pseudo-labels which map the ambiguous representations of pixels to the wrong classes due to the limited cognitive ability of the model. In this paper, we define pixel-wise representations from a new perspective of probability theory and propose a Probabilistic Representation Contrastive Learning (PRCL) framework that improves representation quality by taking its probability into consideration. Through modeling the mapping from pixels to representations as the probability via multivariate Gaussian distributions, we can tune the contribution of the ambiguous representations to tolerate the risk of inaccurate pseudo-labels. Furthermore, we define prototypes in the form of distributions, which indicates the confidence of a class, while the point prototype cannot. Moreover, we propose to regularize the distribution variance to enhance the reliability of representations. Taking advantage of these benefits, high-quality feature representations can be derived in the latent space, thereby the performance of semantic segmentation can be further improved. We conduct sufficient experiment to evaluate PRCL on Pascal VOC and CityScapes. The comparisons with state-of-the-art approaches demonstrate the superiority of proposed PRCL.
翻译:近些年来在半监督的语义分解方面的突破是通过对比性学习开发的。在流行的像素式对比性学习解决方案中,示范地图像素可以确定表层的表达方式,并在潜在空间将其正规化。然而,由于模型认知能力有限,存在着不准确的假标签,可以将象素混混地表示到错误的类别。在本文中,我们从概率理论的新角度来定义象素表示方式,并提出一个概率代表对比学习(PRCL)框架,通过考虑概率来提高代表性质量。通过模拟从像素到表层的映射,通过多变戈斯分布,我们可以调整模模糊的表示方式的贡献,以容忍不准确的假标签的风险。此外,我们用分布方式界定了原型,这表明了某类的信心,而点原型则无法。我们提议调整分布差异,以提高表层的可靠性。利用这些好处,高品质的特征展示可以在潜在空间中进行,从而通过多变式戈斯分布式分布式分布式的概率分析,从而展示了CLR-CR-CLS的状态的演化表现。我们CR-CR-Cal-Calal-Calalalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsalsssalsalsalsalsalsalsssalsalsalsal)。