DWANet: Focus on Foreground Features for More Accurate Location

Keyphrases

• Object Detection 100%
• Accurate Location 100%
• Foreground Features 100%
• Feature Fusion 20%
• Knowledge-how 20%
• Detection Accuracy 20%
• Attention Mechanism 20%
• Bounding Box 20%
• Image Object 20%
• Image Understanding 20%
• Convolutional Neural Network 20%
• Module Base 20%
• Faster R-CNN 20%
• Feature Extraction Tool 20%
• RefineDet 20%
• COCO Dataset 20%
• Effective Feature 20%
• VOC Dataset 20%
• Deformable Convolution 20%

Computer Science

• Object Detection 100%
• Convolutional Neural Network 50%
• Detection Accuracy 25%
• Image Understanding 25%
• Attention (Machine Learning) 25%
• Feature Fusion 25%
• Deformable Convolution 25%
• Feature Extraction 25%