In this paper, we learn to classify visual object instances, incrementally and via self-supervision (self-incremental). Our learner observes a single instance at a time, which is then discarded from the dataset. Incremental instance learning is challenging, since longer learning sessions exacerbate forgetfulness, and labeling instances is cumbersome. We overcome these challenges via three contributions: i. We propose VINIL, a self-incremental learner that can learn object instances sequentially, ii. We equip VINIL with self-supervision to by-pass the need for instance labelling, iii. We compare VINIL to label-supervised variants on two large-scale benchmarks and show that VINIL significantly improves accuracy while reducing forgetfulness.

翻译：在本文中，我们通过自监督（自增量）方式逐步学习视觉对象实例的分类。我们的学习器每次仅观察单个实例，随后该实例从数据集中移除。增量实例学习具有挑战性，因为更长的学习阶段会加剧遗忘问题，且对实例进行标注十分繁琐。我们通过三项贡献克服了这些挑战：i. 提出VINIL，一种能够顺序学习对象实例的自增量学习器；ii. 为VINIL配备自监督机制，从而绕开实例标注的需求；iii. 在两个大规模基准测试上将VINIL与标签监督变体进行比较，结果表明VINIL在降低遗忘的同时显著提升了准确率。