[论文解读] Cell R-CNN V3: A Novel Panoptic Paradigm for Instance Segmentation in Biomedical Images.

Dongnan Liu, Donghao Zhang+3 more|arXiv (Cornell University)|Feb 15, 2020

AI in cancer detection参考文献 32被引用 11

一句话总结

本文提出 Cell R-CNN V3，一种新颖的全景分割框架，用于生物医学图像实例分割，通过残差注意力融合与一致性正则化统一语义与实例特征，减少信息丢失并提升掩码质量预测，在多个生物医学数据集上实现最先进性能。

ABSTRACT

Instance segmentation is an important task for biomedical image analysis. Due to the complicated background components, the high variability of object appearances, numerous overlapping objects, and ambiguous object boundaries, this task still remains challenging. Recently, deep learning based methods have been widely employed to solve these problems and can be categorized into proposal-free and proposal-based methods. However, both proposal-free and proposal-based methods suffer from information loss, as they focus on either global-level semantic or local-level instance features. To tackle this issue, we present a panoptic architecture that unifies the semantic and instance features in this work. Specifically, our proposed method contains a residual attention feature fusion mechanism to incorporate the instance prediction with the semantic features, in order to facilitate the semantic contextual information learning in the instance branch. Then, a mask quality branch is designed to align the confidence score of each object with the quality of the mask prediction. Furthermore, a consistency regularization mechanism is designed between the semantic segmentation tasks in the semantic and instance branches, for the robust learning of both tasks. Extensive experiments demonstrate the effectiveness of our proposed method, which outperforms several state-of-the-art methods on various biomedical datasets.

研究动机与目标

• 为解决在生物医学图像中因仅关注全局语义或局部实例特征而导致的信息丢失挑战。
• 通过专门的质量分支将置信度分数与实际掩码质量对齐，改进掩码质量预测。
• 通过语义分支与实例分支之间的的一致性正则化，提升语义与实例分割任务的鲁棒性。
• 在单一全景架构中统一语义与实例特征学习，以提升在复杂生物医学图像数据上的性能。

提出的方法

• 引入残差注意力特征融合机制，将语义特征融入实例分支，以增强上下文理解。
• 采用掩码质量分支，对每个预测掩码回归一个质量分数，使置信度与实际分割精度对齐。
• 在语义分割头之间应用一致性正则化机制，以稳定并提升语义与实例分支的联合学习。
• 设计统一的全景架构，联合优化实例分割与语义分割任务。
• 利用带有注意力融合的特征金字塔网络，以增强多尺度特征表示。

实验结果

研究问题

• RQ1 统一的全景架构是否能通过在生物医学实例分割中整合语义与实例特征，减少信息丢失？
• RQ2 引入掩码质量分支在多大程度上提升了实例分割预测的可靠性？
• RQ3 语义分支与实例分支之间的一致性正则化在多大程度上增强了模型的鲁棒性与性能？
• RQ4 所提出方法是否在多样化生物医学图像数据集上优于现有最先进方法？

主要发现

• 所提出的 Cell R-CNN V3 在多种生物医学数据集上相较于多个最先进方法表现出更优性能。
• 残差注意力特征融合机制通过有效结合全局语义与局部实例线索，提升了特征表示能力。
• 掩码质量分支通过将置信度分数与实际掩码 IoU 值对齐，增强了预测的可靠性。
• 一致性正则化机制使语义与实例分割任务的联合学习更加稳定且准确。