Multi-view feature High-order Fusion for Space Weak Object Detection and Segmentation

2026-06-13

AI Summary

Weak targets in spatial imagery and video are challenging to represent and detect due to their scarce appearance information. To address this, this work proposes a Multi-view High-order Fusion (MHF) method that leverages a multi-view attention mechanism to generate rich feature representations and introduces a recursive task-contribution gating mechanism to enable high-order feature awareness and adaptive fusion. Designed as a plug-and-play module with strong compatibility, MHF extends conventional low-order fusion to the high-order regime, substantially enhancing weak target modeling capabilities. Evaluated on three space science and satellite video datasets, the proposed approach achieves state-of-the-art performance in both detection and segmentation tasks and consistently improves the efficacy of various mainstream models.

Abstract

Weak objects are common in images and videos of space applications. However, it is hard to learn proper representations from their limited appearance information. Inspired by multi-view learning, we develop simple multi-view attentions, treating their outputs as multi-view features. We also propose a multi-view feature high-order fusion method (MHF) to aggregate more accurate and richer features of weak objects. Our MHF extends the commonly used low-order feature fusion method to higher orders. It enhances the model's capacity to capture relevant and complementary information about weak objects. This is achieved by introducing high-order multi-view features perception and a recursive task-contribution gated selection of multi-view features. The new operation is highly flexible and customizable. It is compatible with various variants of multi-view feature representations. We conduct extensive experiments on two newly constructed space science datasets and an open, large-scale satellite video dataset. Our MHF serves as a plug-and-play module and significantly improves various vision transformers and convolution-based detection and segmentation models. We achieve all state-of-the-art accuracies on both tasks across three datasets. Our MHF can be a new basic module for visual modeling that effectively represents weak objects in terms of multi-view learning. The code will be available at https://github.com/Kingdroper/MHF.

Problem

• weak object
• space imagery
• detection and segmentation
• multi-view learning
• feature representation

Innovation

• multi-view learning
• high-order fusion
• weak object detection
• feature aggregation
• vision transformers