Compressed Meta-Optical Encoder for Image Classification
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916286876876800 |
|---|---|
| author | Wirth-Singh, Anna Xiang, Jinlin Choi, Minho Fröch, Johannes E. Huang, Luocheng Colburn, Shane Shlizerman, Eli Majumdar, Arka |
| author_facet | Wirth-Singh, Anna Xiang, Jinlin Choi, Minho Fröch, Johannes E. Huang, Luocheng Colburn, Shane Shlizerman, Eli Majumdar, Arka |
| contents | Optical and hybrid convolutional neural networks (CNNs) recently have become of increasing interest to achieve low-latency, low-power image classification and computer vision tasks. However, implementing optical nonlinearity is challenging, and omitting the nonlinear layers in a standard CNN comes at a significant reduction in accuracy. In this work, we use knowledge distillation to compress modified AlexNet to a single linear convolutional layer and an electronic backend (two fully connected layers). We obtain comparable performance to a purely electronic CNN with five convolutional layers and three fully connected layers. We implement the convolution optically via engineering the point spread function of an inverse-designed meta-optic. Using this hybrid approach, we estimate a reduction in multiply-accumulate operations from 17M in a conventional electronic modified AlexNet to only 86K in the hybrid compressed network enabled by the optical frontend. This constitutes over two orders of magnitude reduction in latency and power consumption. Furthermore, we experimentally demonstrate that the classification accuracy of the system exceeds 93% on the MNIST dataset. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_06534 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Compressed Meta-Optical Encoder for Image Classification Wirth-Singh, Anna Xiang, Jinlin Choi, Minho Fröch, Johannes E. Huang, Luocheng Colburn, Shane Shlizerman, Eli Majumdar, Arka Computer Vision and Pattern Recognition Image and Video Processing Optics Optical and hybrid convolutional neural networks (CNNs) recently have become of increasing interest to achieve low-latency, low-power image classification and computer vision tasks. However, implementing optical nonlinearity is challenging, and omitting the nonlinear layers in a standard CNN comes at a significant reduction in accuracy. In this work, we use knowledge distillation to compress modified AlexNet to a single linear convolutional layer and an electronic backend (two fully connected layers). We obtain comparable performance to a purely electronic CNN with five convolutional layers and three fully connected layers. We implement the convolution optically via engineering the point spread function of an inverse-designed meta-optic. Using this hybrid approach, we estimate a reduction in multiply-accumulate operations from 17M in a conventional electronic modified AlexNet to only 86K in the hybrid compressed network enabled by the optical frontend. This constitutes over two orders of magnitude reduction in latency and power consumption. Furthermore, we experimentally demonstrate that the classification accuracy of the system exceeds 93% on the MNIST dataset. |
| title | Compressed Meta-Optical Encoder for Image Classification |
| topic | Computer Vision and Pattern Recognition Image and Video Processing Optics |
| url | https://arxiv.org/abs/2406.06534 |