SOTAVerified

RGB-T Salient Object Detection

RGB-T Salient Object Detection (SOD) focuses on identifying the most visually prominent objects or regions in a scene using both RGB (color) and thermal imaging data. This technique leverages the complementary strengths of visible and thermal spectrums to enhance the detection of salient objects, particularly useful in challenging visibility conditions such as night, fog, or smoke. Applications of RGB-T SOD are broad and include video/image segmentation, object recognition, visual tracking, and enhanced surveillance systems. Thermal data particularly aids in scenarios where color and texture information is insufficient for accurate detection. This method is also valuable in search and rescue operations, wildlife monitoring, and advanced driver-assistance systems (ADAS). Online benchmark and resources can be accessed at a dedicated platform for further research and development in this area.

Papers

Showing 110 of 10 papers

TitleStatusHype
KAN-SAM: Kolmogorov-Arnold Network Guided Segment Anything Model for RGB-T Salient Object Detection0
Alignment-Free RGB-T Salient Object Detection: A Large-scale Dataset and Progressive Correlation NetworkCode1
Divide-and-Conquer: Confluent Triple-Flow Network for RGB-T Salient Object Detection0
CoLA: Conditional Dropout and Language-driven Robust Dual-modal Salient Object DetectionCode1
All in One: RGB, RGB-D, and RGB-T Salient Object Detection0
Scribble-Supervised RGB-T Salient Object DetectionCode1
Interactive Context-Aware Network for RGB-T Salient Object Detection0
Does Thermal Really Always Matter for RGB-T Salient Object Detection?Code0
SwinNet: Swin Transformer drives edge-aware RGB-D and RGB-T salient object detectionCode1
CAVER: Cross-Modal View-Mixed Transformer for Bi-Modal Salient Object DetectionCode1
Show:102550

No leaderboard results yet.