AlignSAM: Aligning Segment Anything Model to Open Context via Reinforcement Learning

TL;DR

AlignSAM introduces a reinforcement learning-based framework that automatically generates prompts to adapt the Segment Anything Model to diverse open-world segmentation tasks without retraining, improving accuracy and generalization.

Contribution

This paper presents a novel reinforcement learning approach for automatic prompting, enabling SAM to adapt to various tasks while keeping its parameters frozen, which is a significant advancement over existing methods.

Findings

AlignSAM outperforms state-of-the-art methods on multiple segmentation benchmarks.

The reinforcement learning policy effectively generates prompts that improve segmentation accuracy.

The semantic recalibration module enhances handling of explicit and implicit semantics.

Abstract

Powered by massive curated training data, Segment Anything Model (SAM) has demonstrated its impressive generalization capabilities in open-world scenarios with the guidance of prompts. However, the vanilla SAM is class agnostic and heavily relies on user-provided prompts to segment objects of interest. Adapting this method to diverse tasks is crucial for accurate target identification and to avoid suboptimal segmentation results. In this paper, we propose a novel framework, termed AlignSAM, designed for automatic prompting for aligning SAM to an open context through reinforcement learning. Anchored by an agent, AlignSAM enables the generality of the SAM model across diverse downstream tasks while keeping its parameters frozen. Specifically, AlignSAM initiates a prompting agent to iteratively refine segmentation predictions by interacting with the foundational model. It integrates a reinforcement learning policy network to provide informative prompts to the foundational models. Additionally, a semantic recalibration module is introduced to provide fine-grained labels of prompts, enhancing the model's proficiency in handling tasks encompassing explicit and implicit semantics. Experiments conducted on various challenging segmentation tasks among existing foundation models demonstrate the superiority of the proposed AlignSAM over state-of-the-art approaches. Project page: \url{https://github.com/Duojun-Huang/AlignSAM-CVPR2024}.