面向类别增量语义分割中的真实增量场景
Towards Realistic Incremental Scenario in Class Incremental Semantic Segmentation
AI总结：
本文指出连续增量语义分割的重叠式场景存在不真实问题且会导致技术评估偏差，提出分区式实用场景及样本记忆数据检索的代码修复方案，还提出基线方法MiB-AugM，在多类新类别学习任务中取得最优结果。
摘要：
本文针对常用的连续增量语义分割（Continuous Incremental Semantic Segmentation，CISS）场景中被称为“重叠式（overlapped）”的不真实问题展开研究。我们指出，重叠式场景允许同一张图像在后续任务中以不同像素标签重新出现，这与实际的增量学习场景相去甚远。此外，我们发现这种存在缺陷的场景可能导致CISS中两种常用技术——伪标签（pseudo-labeling）和样本记忆（exemplar memory）——产生有偏差的结果，给某些技术带来非预期的优势或劣势。为缓解该问题，本文提出了一种名为“分区式（partitioned）”的实用场景：首先将数据集划分为对应各个类别的不同子集，再将这些子集分配给对应的任务。该方案在满足CISS场景（如捕捉背景偏移）要求的同时，有效解决了上述问题。此外，我们还发现并解决了以往研究中被忽略的、与从样本记忆中检索数据相关的代码实现问题。最后，我们提出了一种简单但具有竞争力的基于记忆的基线方法MiB-AugM，该方法可在样本记忆中处理当前任务的背景偏移。该基线在涉及大量新类别学习的多项任务中取得了当前最优（state-of-the-art）结果。
英文摘要：
This paper addresses the unrealistic aspect of the commonly adopted Continuous Incremental Semantic Segmentation (CISS) scenario, termed overlapped. We point out that overlapped allows the same image to reappear in future tasks with different pixel labels, which is far from practical incremental learning scenarios. Moreover, we identified that this flawed scenario may lead to biased results for two commonly used techniques in CISS, pseudo-labeling and exemplar memory, resulting in unintended advantages or disadvantages for certain techniques. To mitigate this, a practical scenario called partitioned is proposed, in which the dataset is first divided into distinct subsets representing each class, and then the subsets are assigned to each corresponding task. This efficiently addresses the issue above while meeting the requirement of CISS scenario, such as capturing the background shifts. Furthermore, we identify and address the code implementation issues related to retrieving data from the exemplar memory, which was ignored in previous works. Lastly, we introduce a simple yet competitive memory-based baseline, MiB-AugM, that handles background shifts of current tasks in the exemplar memory. This baseline achieves state-of-the-art results across multiple tasks involving learning numerous new classes.
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。
