Self-supervised Learning of Contextualized Local Visual Embeddings

TL;DR

This paper introduces CLoVE, a self-supervised method that learns contextualized local visual embeddings using a novel attention mechanism, achieving state-of-the-art results in dense prediction tasks.

Contribution

The paper proposes a new self-supervised learning approach with a normalized multi-head self-attention layer for dense visual representations, outperforming existing methods.

Findings

State-of-the-art performance in object detection

Superior results in instance segmentation

Effective in keypoint detection and dense pose estimation

Abstract

We present Contextualized Local Visual Embeddings (CLoVE), a self-supervised convolutional-based method that learns representations suited for dense prediction tasks. CLoVE deviates from current methods and optimizes a single loss function that operates at the level of contextualized local embeddings learned from output feature maps of convolution neural network (CNN) encoders. To learn contextualized embeddings, CLoVE proposes a normalized mult-head self-attention layer that combines local features from different parts of an image based on similarity. We extensively benchmark CLoVE's pre-trained representations on multiple datasets. CLoVE reaches state-of-the-art performance for CNN-based architectures in 4 dense prediction downstream tasks, including object detection, instance segmentation, keypoint detection, and dense pose estimation.