Supplementary Material for Self-Supervised Visual Representation Learning with Semantic Grouping Xin Wen

There are two operations in our data augmentation pipeline that changes the scale or layout of the image, i.e ., random resized crop and random horizontal flip. This is followed by a resize operation to recover the intersect part to the original size ( e.g ., RoIAlign to recover the original spatial layout. The total stride is 16 (FCN-16s [20]). Intuitively, each prototype can be viewed as the cluster center of a semantic class. During inference, we only take the teacher model parameterized by ξ .