RGBT Salient Object Detection: A Large-scale Dataset and Benchmark
share
Salient object detection in complex scenes and environments is a challenging research topic. limits its performance of real-life applications when confronted with adverse conditions such as dark environments and complex backgrounds. advantage of RGB and thermal infrared images becomes a new research direction for detecting salient object in complex scenes recently, as thermal infrared spectrum imaging provides the complementary information and has been applied to many computer vision tasks. detection is limited by the lack of a large-scale dataset and comprehensive benchmark. including 5000 spatially aligned RGBT image pairs with ground truth annotations. environments for exploring the robustness of algorithms. we propose a powerful baseline approach, which extracts multi-level features within each modality and aggregates these features of all modalities with the attention mechanism, for accurate RGBT salient object detection. experiments show that the proposed baseline approach outperforms the state-of-the-art methods on VT5000 dataset and other two public datasets. addition, we carry out a comprehensive analysis of different algorithms of RGBT salient object detection on VT5000 dataset, and then make several valuable conclusions and provide some potential research directions for RGBT salient object detection.
READ FULL TEXT