Answer-first summary for fast verification
Answer: Use a lightweight model architecture, label with bounding boxes, evaluate using mean average precision.
For efficient and accurate object detection, a lightweight model architecture is preferred to ensure real-time performance. Labeling with bounding boxes is necessary for object detection models. Evaluating using mean average precision (mAP) is appropriate as it provides a comprehensive measure of how well the model performs across different IoU thresholds, which is crucial for object detection tasks.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
You have been asked to train a custom object detection model for a smart city project that needs to detect and count various types of vehicles in real-time traffic footage. The model needs to be efficient and accurate. Describe how you would approach the training of this model, including the choice of model architecture, the labeling process, and the metrics you would use to evaluate the model's performance.
A
Use a lightweight model architecture, label with object tags, evaluate using recall.
B
Use a complex model architecture, label with bounding boxes, evaluate using mean average precision.
C
Use a lightweight model architecture, label with bounding boxes, evaluate using mean average precision.
D
Use a complex model architecture, label with object tags, evaluate using accuracy.