Comparative Study of Parameter Selection for Enhanced Edge Inference for a Multi-Output Regression model for Head Pose Estimation
arXiv:2302.00592v1h-index: 12
Originality Synthesis-oriented
AI Analysis
This work addresses the challenge of deploying efficient models on edge devices for head pose estimation, but it appears incremental as it focuses on parameter selection and pruning techniques.
The study tackled the problem of optimizing deep learning models for edge inference by applying magnitude-based pruning to a multi-output regression model for head pose estimation, achieving over 75% model size reduction with higher accuracy than the original model.
Magnitude-based pruning is a technique used to optimise deep learning models for edge inference. We have achieved over 75% model size reduction with a higher accuracy than the original multi-output regression model for head-pose estimation.