Unified-modal Salient Object Detection via Adaptive Prompt Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Kunpeng, Li, Chenglong, Tu, Zhengzheng, Liu, Zhengyi, Luo, Bin
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916274091589632
author Wang, Kunpeng
Li, Chenglong
Tu, Zhengzheng
Liu, Zhengyi
Luo, Bin
author_facet Wang, Kunpeng
Li, Chenglong
Tu, Zhengzheng
Liu, Zhengyi
Luo, Bin
contents Existing single-modal and multi-modal salient object detection (SOD) methods focus on designing specific architectures tailored for their respective tasks. However, developing completely different models for different tasks leads to labor and time consumption, as well as high computational and practical deployment costs. In this paper, we attempt to address both single-modal and multi-modal SOD in a unified framework called UniSOD, which fully exploits the overlapping prior knowledge between different tasks. Nevertheless, assigning appropriate strategies to modality variable inputs is challenging. To this end, UniSOD learns modality-aware prompts with task-specific hints through adaptive prompt learning, which are plugged into the proposed pre-trained baseline SOD model to handle corresponding tasks, while only requiring few learnable parameters compared to training the entire model. Each modality-aware prompt is generated from a switchable prompt generation block, which adaptively performs structural switching based on single-modal and multi-modal inputs without human intervention. Through end-to-end joint training, UniSOD achieves overall performance improvement on 14 benchmark datasets for RGB, RGB-D, and RGB-T SOD, which demonstrates that our method effectively and efficiently unifies single-modal and multi-modal SOD tasks.The code and results are available at https://github.com/Angknpng/UniSOD.
format Preprint
id arxiv_https___arxiv_org_abs_2311_16835
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Unified-modal Salient Object Detection via Adaptive Prompt Learning
Wang, Kunpeng
Li, Chenglong
Tu, Zhengzheng
Liu, Zhengyi
Luo, Bin
Computer Vision and Pattern Recognition
Existing single-modal and multi-modal salient object detection (SOD) methods focus on designing specific architectures tailored for their respective tasks. However, developing completely different models for different tasks leads to labor and time consumption, as well as high computational and practical deployment costs. In this paper, we attempt to address both single-modal and multi-modal SOD in a unified framework called UniSOD, which fully exploits the overlapping prior knowledge between different tasks. Nevertheless, assigning appropriate strategies to modality variable inputs is challenging. To this end, UniSOD learns modality-aware prompts with task-specific hints through adaptive prompt learning, which are plugged into the proposed pre-trained baseline SOD model to handle corresponding tasks, while only requiring few learnable parameters compared to training the entire model. Each modality-aware prompt is generated from a switchable prompt generation block, which adaptively performs structural switching based on single-modal and multi-modal inputs without human intervention. Through end-to-end joint training, UniSOD achieves overall performance improvement on 14 benchmark datasets for RGB, RGB-D, and RGB-T SOD, which demonstrates that our method effectively and efficiently unifies single-modal and multi-modal SOD tasks.The code and results are available at https://github.com/Angknpng/UniSOD.
title Unified-modal Salient Object Detection via Adaptive Prompt Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.16835