GraspSAM: When Segment Anything Model Meets Grasp Detection

TL;DR

GraspSAM is a novel prompt-driven, category-agnostic grasp detection model that leverages the Segment Anything Model's capabilities, achieving state-of-the-art results with minimal fine-tuning across diverse datasets.

Contribution

It introduces GraspSAM, an innovative extension of SAM, integrating object segmentation and grasp prediction with prompt-based control for flexible, data-efficient robotic grasping.

Findings

Achieves state-of-the-art performance on multiple datasets.

Supports various prompts like points, boxes, and language.

Demonstrates robustness in real-world applications.

Abstract

Grasp detection requires flexibility to handle objects of various shapes without relying on prior knowledge of the object, while also offering intuitive, user-guided control. This paper introduces GraspSAM, an innovative extension of the Segment Anything Model (SAM), designed for prompt-driven and category-agnostic grasp detection. Unlike previous methods, which are often limited by small-scale training data, GraspSAM leverages the large-scale training and prompt-based segmentation capabilities of SAM to efficiently support both target-object and category-agnostic grasping. By utilizing adapters, learnable token embeddings, and a lightweight modified decoder, GraspSAM requires minimal fine-tuning to integrate object segmentation and grasp prediction into a unified framework. The model achieves state-of-the-art (SOTA) performance across multiple datasets, including Jacquard, Grasp-Anything, and Grasp-Anything++. Extensive experiments demonstrate the flexibility of GraspSAM in handling different types of prompts (such as points, boxes, and language), highlighting its robustness and effectiveness in real-world robotic applications.