A Joint Analysis of Input Resolution and Quantization Precision in Deep Learning
S. K. Kim, Minjae Kim, Youngki Lee · 2023
Deep learning models have become increasingly prevalent in various domains, necessitating their deployment on resource-constrained devices. Quantization is a promising way to reduce the model complexity in that it keeps model architecture intact and enables the model to operate on specialized hardwares(e.g., NPU, DSP). Input resolution is also essential in making a trade-off between accuracy and computation.