Quantized neural network for complex hologram generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Endo, Yutaka, Oikawa, Minoru, Wilkinson, Timothy D., Shimobaba, Tomoyoshi, Ito, Tomoyoshi
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929570025832448
author Endo, Yutaka
Oikawa, Minoru
Wilkinson, Timothy D.
Shimobaba, Tomoyoshi
Ito, Tomoyoshi
author_facet Endo, Yutaka
Oikawa, Minoru
Wilkinson, Timothy D.
Shimobaba, Tomoyoshi
Ito, Tomoyoshi
contents Computer-generated holography (CGH) is a promising technology for augmented reality displays, such as head-mounted or head-up displays. However, its high computational demand makes it impractical for implementation. Recent efforts to integrate neural networks into CGH have successfully accelerated computing speed, demonstrating the potential to overcome the trade-off between computational cost and image quality. Nevertheless, deploying neural network-based CGH algorithms on computationally limited embedded systems requires more efficient models with lower computational cost, memory footprint, and power consumption. In this study, we developed a lightweight model for complex hologram generation by introducing neural network quantization. Specifically, we built a model based on tensor holography and quantized it from 32-bit floating-point precision (FP32) to 8-bit integer precision (INT8). Our performance evaluation shows that the proposed INT8 model achieves hologram quality comparable to that of the FP32 model while reducing the model size by approximately 70% and increasing the speed fourfold. Additionally, we implemented the INT8 model on a system-on-module to demonstrate its deployability on embedded platforms and high power efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2409_06711
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Quantized neural network for complex hologram generation
Endo, Yutaka
Oikawa, Minoru
Wilkinson, Timothy D.
Shimobaba, Tomoyoshi
Ito, Tomoyoshi
Computer Vision and Pattern Recognition
Graphics
Computer-generated holography (CGH) is a promising technology for augmented reality displays, such as head-mounted or head-up displays. However, its high computational demand makes it impractical for implementation. Recent efforts to integrate neural networks into CGH have successfully accelerated computing speed, demonstrating the potential to overcome the trade-off between computational cost and image quality. Nevertheless, deploying neural network-based CGH algorithms on computationally limited embedded systems requires more efficient models with lower computational cost, memory footprint, and power consumption. In this study, we developed a lightweight model for complex hologram generation by introducing neural network quantization. Specifically, we built a model based on tensor holography and quantized it from 32-bit floating-point precision (FP32) to 8-bit integer precision (INT8). Our performance evaluation shows that the proposed INT8 model achieves hologram quality comparable to that of the FP32 model while reducing the model size by approximately 70% and increasing the speed fourfold. Additionally, we implemented the INT8 model on a system-on-module to demonstrate its deployability on embedded platforms and high power efficiency.
title Quantized neural network for complex hologram generation
topic Computer Vision and Pattern Recognition
Graphics
url https://arxiv.org/abs/2409.06711