Back to course
Next page
Interactive Quiz
Test your knowledge!
1
What is the main difference between a two-stage object detector and a one-stage detector?
A
The two-stage detector directly produces classes and boxes, while the one-stage detector first generates region proposals.
B
The two-stage detector uses a transformer, while the one-stage detector uses only a CNN.
C
The two-stage detector first generates region proposals and then refines the detection, while the one-stage detector directly predicts boxes and classes in a single step.
D
The one-stage detector is always more accurate than the two-stage detector.
2
In the original YOLO architecture, how is the cell responsible for detecting an object determined?
A
The cell containing the center of the object in the grid is responsible for detection.
B
The cell that has the largest area of the object is responsible.
C
All cells predict all boxes and classes without specific assignment.
D
The cell at the top-left corner of the object is responsible.
3
What is the main role of Non-Maximum Suppression (NMS) in object detection?
A
Improve image resolution before detection.
B
Merge multiple overlapping box predictions into a single detection per object.
C
Generate region proposals for detection.
D
Reduce the size of the anchors used by the model.
4
What is the main innovation introduced in YOLOv2 compared to the original YOLO version?
A
Use of a Darknet-53 backbone.
B
Introduction of anchor boxes to improve localization accuracy.
C
Addition of multi-scale detection.
D
Use of a transformer encoder.
5
What is one of the major limitations of the original YOLO algorithm related to grid cell prediction?
A
Each cell can predict only one class but multiple boxes.
B
Each cell can detect multiple objects of different classes.
C
The model does not predict object positions, only classes.
D
Each cell can predict only one box and one class.
Score: 0/5
Score: 0/5