Are Labels Needed for Incremental Instance Learning?
This addresses the challenge of incremental learning for computer vision applications where labeling is cumbersome, though it appears incremental in method.
The paper tackles the problem of incremental instance learning for visual object classification by proposing VINIL, a self-supervised method that learns sequentially without labels, and shows it significantly improves accuracy and reduces forgetfulness compared to supervised variants on large-scale benchmarks.
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.