Intelligent Anomaly Detection for Lane Rendering Using Transformer with Self-Supervised Pre-Training and Customized Fine-Tuning

INIS

• transformers 100%
• images 100%
• tuning 100%
• detection 100%
• pipelines 60%
• accuracy 60%
• maps 40%
• curves 40%
• processing 40%
• performance 40%
• data 40%
• modeling 40%
• ablation 20%
• losses 20%
• masking 20%
• comparative evaluations 20%
• entropy 20%
• navigation 20%
• learning 20%
• humans 20%
• classification 20%

Computer Science

• Anomaly Detection 100%
• Image Modeling 66%
• Image Rendering 66%
• Swin Transformer 66%
• Deep Learning Technique 33%
• Performance Model 33%
• Postprocessing 33%
• Data Preprocessing 33%
• Detection Accuracy 33%