ProMerge: Prompt and Merge for Unsupervised Instance Segmentation

TL;DR

ProMerge introduces a fast and effective method for unsupervised instance segmentation by combining prompt-based grouping and merging strategies, outperforming existing approaches in accuracy and speed.

Contribution

The paper presents ProMerge, a novel approach that reduces inference time and improves segmentation quality using strategic merging and background pruning in an unsupervised setting.

Findings

ProMerge achieves competitive segmentation results.

It significantly reduces inference time compared to normalized-cut methods.

Using ProMerge masks to train detectors surpasses current unsupervised models.

Abstract

Unsupervised instance segmentation aims to segment distinct object instances in an image without relying on human-labeled data. This field has recently seen significant advancements, partly due to the strong local correspondences afforded by rich visual feature representations from self-supervised models (e.g., DINO). Recent state-of-the-art approaches use self-supervised features to represent images as graphs and solve a generalized eigenvalue system (i.e., normalized-cut) to generate foreground masks. While effective, this strategy is limited by its attendant computational demands, leading to slow inference speeds. In this paper, we propose Prompt and Merge (ProMerge), which leverages self-supervised visual features to obtain initial groupings of patches and applies a strategic merging to these segments, aided by a sophisticated background-based mask pruning technique. ProMerge not only yields competitive results but also offers a significant reduction in inference time compared to state-of-the-art normalized-cut-based approaches. Furthermore, when training an object detector using our mask predictions as pseudo-labels, the resulting detector surpasses the current leading unsupervised model on various challenging instance segmentation benchmarks.