Prompt-Free Universal Region Proposal Network

2026-03-18

AI Summary

This work proposes the first universal region proposal network that operates without any external image or text prompts, addressing the limited generalizability of existing methods in real-world scenarios. The approach leverages learnable query embeddings to dynamically localize potential objects and introduces a cascade self-prompting mechanism to iteratively refine region proposals. It integrates a Sparse Image-Aware Adapter, Centerness-Guided Query Selection, and a Cascade Self-Prompt module, enabling effective training with only 5% of the COCO dataset and supporting cross-domain zero-shot transfer. Extensive experiments across 19 diverse datasets—including underwater imagery, industrial defect detection, and remote sensing—demonstrate the method’s superior performance and broad applicability.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Identifying potential objects is critical for object recognition and analysis across various computer vision applications. Existing methods typically localize potential objects by relying on exemplar images, predefined categories, or textual descriptions. However, their reliance on image and text prompts often limits flexibility, restricting adaptability in real-world scenarios. In this paper, we introduce a novel Prompt-Free Universal Region Proposal Network (PF-RPN), which identifies potential objects without relying on external prompts. First, the Sparse Image-Aware Adapter (SIA) module performs initial localization of potential objects using a learnable query embedding dynamically updated with visual features. Next, the Cascade Self-Prompt (CSP) module identifies the remaining potential objects by leveraging the self-prompted learnable embedding, autonomously aggregating informative visual features in a cascading manner. Finally, the Centerness-Guided Query Selection (CG-QS) module facilitates the selection of high-quality query embeddings using a centerness scoring network. Our method can be optimized with limited data (e.g., 5% of MS COCO data) and applied directly to various object detection application domains for identifying potential objects without fine-tuning, such as underwater object detection, industrial defect detection, and remote sensing image object detection. Experimental results across 19 datasets validate the effectiveness of our method. Code is available at https://github.com/tangqh03/PF-RPN.

Problem

• region proposal
• prompt-free
• object detection
• universal object localization
• computer vision

Innovation

• Prompt-Free
• Universal Region Proposal
• Self-Prompting
• Query Embedding
• Cross-Domain Generalization