Low-Latency INT8 Quantized Object Detection on Embedded Jetson
Objective & Abstract
Profiling post-training calibration loss and FPS throughput of YOLOv8n on resource-constrained ARM/NVIDIA architectures with layer-wise sensitivity audits.
Investigated Methodologies
- Post-Training Quantization (PTQ)
- TensorRT layer fusion
- Thermal power profiling
Datasets & Benchmarking Corpora
COCO 2017 Validation Subset
Campus Edge Telemetry Dataset
All training and evaluation datasets undergo ethics audit and licensing compliance per AIAC Research Council standards.