Ground Truth
الحقيقة الأرضية
التسميات الصحيحة المعتمدة يدويًّا (مربعات الإحاطة وتصنيفات الفئات) التي يُقاس أداء النموذج مقابلها.
The manually annotated correct labels (bounding boxes and class labels) against which a model's predictions are evaluated.
Also translated asالبيانات المرجعية الحقيقية، التسمية المرجعية
First appears in this corpus in: Microsoft COCO: Common Objects in Context (2014)
Appears in these papers
- beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework2017in the sky ✦
- Microsoft COCO: Common Objects in Context2014in the sky ✦
- DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs2017in the sky ✦
- Fast R-CNN2015in the sky ✦
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks2015in the sky ✦
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014in the sky ✦
- Focal Loss for Dense Object Detection2017in the sky ✦
- Segment Anything2023in the sky ✦
- SSD: Single Shot MultiBox Detector2016in the sky ✦
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015in the sky ✦
- VQA: Visual Question Answering2015in the sky ✦
- You Only Look Once: Unified, Real-Time Object Detection2015in the sky ✦
- YOLOv3: An Incremental Improvement2018in the sky ✦