MSRNet: A Multi-Scale Recursive Network for Camouflaged Object Detection
Leena Alghamdi, Muhammad Usman, Hafeez Anwar, Abdul Bais, Saeed Anwar
2025-11-25
Summary
This paper focuses on the difficult problem of finding objects that are really good at hiding – things that blend into their surroundings so well they're hard to see, like an insect on a leaf. It introduces a new computer vision method to improve how well computers can detect these camouflaged objects.
What's the problem?
Detecting camouflaged objects is tough because they look so similar to what's around them in terms of color and texture. It gets even harder when the lighting is bad, parts of the object are hidden, the object is small, the background is busy, or there are many camouflaged objects all at once. Existing methods aren't very accurate, especially when dealing with small or multiple hidden objects, meaning there's a need for better techniques.
What's the solution?
The researchers created a new system called a Multi-Scale Recursive Network. This system works by looking at the image at different levels of detail, kind of like zooming in and out. It then smartly combines these different views using something called 'attention,' focusing on the most important parts. Finally, it refines its detection through a process of repeated improvement, using information from the whole image to understand the context and pinpoint the hidden objects more accurately.
Why it matters?
This research is important because it significantly improves the ability of computers to find camouflaged objects. This has real-world applications in areas like environmental monitoring (finding hidden animals), security (detecting disguised threats), and even quality control (spotting defects that blend in). The new method achieves top results on standard tests, showing it's a major step forward in this challenging field.
Abstract
Camouflaged object detection is an emerging and challenging computer vision task that requires identifying and segmenting objects that blend seamlessly into their environments due to high similarity in color, texture, and size. This task is further complicated by low-light conditions, partial occlusion, small object size, intricate background patterns, and multiple objects. While many sophisticated methods have been proposed for this task, current methods still struggle to precisely detect camouflaged objects in complex scenarios, especially with small and multiple objects, indicating room for improvement. We propose a Multi-Scale Recursive Network that extracts multi-scale features via a Pyramid Vision Transformer backbone and combines them via specialized Attention-Based Scale Integration Units, enabling selective feature merging. For more precise object detection, our decoder recursively refines features by incorporating Multi-Granularity Fusion Units. A novel recursive-feedback decoding strategy is developed to enhance global context understanding, helping the model overcome the challenges in this task. By jointly leveraging multi-scale learning and recursive feature optimization, our proposed method achieves performance gains, successfully detecting small and multiple camouflaged objects. Our model achieves state-of-the-art results on two benchmark datasets for camouflaged object detection and ranks second on the remaining two. Our codes, model weights, and results are available at https://github.com/linaagh98/MSRNet{https://github.com/linaagh98/MSRNet}.