Scholarlink

<p>Artificial neural networks have fundamentally transformed the field of computer vision, providing unprecedented performance. However, these neural networks for image processing demand substantial computational resources, often hindering real-time operation. In this work, we demonstrate an optical encoder that can perform convolution simultaneously in three color channels during the image capture, effectively implementing several initial convolutional layers of the network. Such an optical encoding results in &#xa0;~&#xa0;24,&#xa0;000&#xa0;× reduction in computational operations, with a state-of-the-art classification accuracy (~73.2%) in free-space optical system. In addition, our analog optical encoder, trained for CIFAR-10 data, can be transferred to the ImageNet subset, High-10, without any modifications, and still exhibits moderate accuracy. The proposed method can decrease total system-level energy more than two orders of magnitude per a single object classification. Our results evidence the potential of hybrid optical/digital computer vision system in which the optical frontend can pre-process an ambient scene to reduce the energy and latency of the whole computer vision system.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Transferable polychromatic optical encoder for neural networks

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月1日

查看全文
  • Minho Choi,
  • Jinlin Xiang,
  • Anna Wirth-Singh,
  • Seung-Hwan Baek,
  • Eli Shlizerman,
  • Arka Majumdar

摘要

Artificial neural networks have fundamentally transformed the field of computer vision, providing unprecedented performance. However, these neural networks for image processing demand substantial computational resources, often hindering real-time operation. In this work, we demonstrate an optical encoder that can perform convolution simultaneously in three color channels during the image capture, effectively implementing several initial convolutional layers of the network. Such an optical encoding results in  ~ 24, 000 × reduction in computational operations, with a state-of-the-art classification accuracy (~73.2%) in free-space optical system. In addition, our analog optical encoder, trained for CIFAR-10 data, can be transferred to the ImageNet subset, High-10, without any modifications, and still exhibits moderate accuracy. The proposed method can decrease total system-level energy more than two orders of magnitude per a single object classification. Our results evidence the potential of hybrid optical/digital computer vision system in which the optical frontend can pre-process an ambient scene to reduce the energy and latency of the whole computer vision system.