Green Video Camouflaged Object Detection

TL;DR

GreenVCOD introduces a resource-efficient video camouflaged object detection method that leverages temporal neighborhoods for improved context understanding, achieving competitive results with reduced complexity.

Contribution

It proposes a novel green VCOD approach using temporal neighborhoods to enhance detection while maintaining low complexity.

Findings

GreenVCOD achieves competitive performance on VCOD benchmarks.

The method effectively captures joint spatial and temporal context.

It demonstrates improved efficiency over existing methods.

Abstract

Camouflaged object detection (COD) aims to distinguish hidden objects embedded in an environment highly similar to the object. Conventional video-based COD (VCOD) methods explicitly extract motion cues or employ complex deep learning networks to handle the temporal information, which is limited by high complexity and unstable performance. In this work, we propose a green VCOD method named GreenVCOD. Built upon a green ICOD method, GreenVCOD uses long- and short-term temporal neighborhoods (TN) to capture joint spatial/temporal context information for decision refinement. Experimental results show that GreenVCOD offers competitive performance compared to state-of-the-art VCOD benchmarks.