Explain the trade-offs between different object detection architectures (e.g., YOLO vs. Faster R-CNN) for Qualcomm mobile and edge vision workloads.
