SPLF-SAM: Self-Prompting Segment Anything Model for Light Field Salient Object Detection

TL;DR

This paper introduces SPLF-SAM, a novel light field salient object detection model that leverages self-prompting, multi-scale feature embedding, and frequency analysis to improve detection accuracy, especially for small objects.

Contribution

The paper proposes SPLF-SAM with UMFEB and MAFA, integrating multi-scale features and frequency-domain analysis to enhance light field salient object detection.

Findings

Outperforms ten state-of-the-art LF SOD methods

Effectively detects small objects amidst noise

Utilizes frequency features for improved accuracy

Abstract

Segment Anything Model (SAM) has demonstrated remarkable capabilities in solving light field salient object detection (LF SOD). However, most existing models tend to neglect the extraction of prompt information under this task. Meanwhile, traditional models ignore the analysis of frequency-domain information, which leads to small objects being overwhelmed by noise. In this paper, we put forward a novel model called self-prompting light field segment anything model (SPLF-SAM), equipped with unified multi-scale feature embedding block (UMFEB) and a multi-scale adaptive filtering adapter (MAFA). UMFEB is capable of identifying multiple objects of varying sizes, while MAFA, by learning frequency features, effectively prevents small objects from being overwhelmed by noise. Extensive experiments have demonstrated the superiority of our method over ten state-of-the-art (SOTA) LF SOD methods. Our code will be available at https://github.com/XucherCH/splfsam.