Abstract
Recently, deep learning-based facial landmark detection has achieved significant improvement. However, the semantic ambiguity problem degrades detection performance. Specifically, the semantic ambiguity causes inconsistent annotation and negatively affects the model's convergence, leading to worse accuracy and instability prediction. To solve this problem, we propose a Self-adapTive Ambiguity Reduction (STAR) loss by exploiting the properties of semantic ambiguity. We find that semantic ambiguity results in the anisotropic predicted distribution, which inspires us to use predicted distribution to represent semantic ambiguity. Based on this, we design the STAR loss that measures the anisotropism of the predicted distribution. Compared with the standard regression loss, STAR loss is encouraged to be small when the predicted distribution is anisotropic and thus adaptively mitigates the impact of semantic ambiguity. Moreover, we propose two kinds of eigenvalue restriction methods that could avoid both distribution's abnormal change and the model's premature convergence. Finally, the comprehensive experiments demonstrate that STAR loss outperforms the state-of-the-art methods on three benchmarks, i.e., COFW, 300W, and WFLW, with negligible computation overhead. Code is at this https URL.
Abstract (translated)
最近,基于深度学习的面部地标检测取得了显著改进。然而,语义歧义问题削弱了检测性能。具体来说,语义歧义导致不一致的标注并负面影响模型收敛,导致更准确的预测但不稳定的预测。为了解决这一问题,我们提出了一种自适应歧义减少(STAR)损失,利用语义歧义的特性。我们发现,语义歧义导致预测分布anisotropic,启发我们使用预测分布来表示语义歧义。基于这种情况,我们设计了STAR损失,用于衡量预测分布的异质性。与传统的回归损失相比,STAR损失在预测分布anisotropic时Encouraged to be small,从而自适应地减缓语义歧义的影响。此外,我们提出了两种特征向量限制方法,可以避免分布的异常变化和模型的过早收敛。最终,综合实验表明,STAR损失在三个基准问题上(COFW、300W和WFLW)优于最先进的方法,并且计算开销为零。代码在此https URL。
URL
https://arxiv.org/abs/2306.02763
