一比特足矣！基于单比特成像的高效视频采集
A Bit is All You Need! Efficient Video Capture via Single Bit Imaging
把传感器端采集压缩到 1 bit（时变阈值量化），再用神经网络重建全比特深度视频，为超低功耗/高速相机流水线提供了一条可复用的「采集-重建」联合设计思路，可直接迁移到事件相机、高速视频与压缩成像任务。
原文依据 one bit measurements captured by time-varying thresholding can be used to reconstruct full-bit-depth videos
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出了一种全新的视频感知范式——1 比特计算视频，它重新定义了成像效率与性能的极限。与传统的高比特深度采集不同，我们证明由时变阈值化获得的单比特测量可用于重建全比特深度视频，从而无需在传感器端使用功耗高昂的高精度模数转换器（ADC），同时降低数据传输中的能耗。我们提出了阈值化策略，以有效捕获视频流中的时空依赖关系。尽管在采集阶段实现了显著的数据压缩，我们仍通过神经视频重建以高保真度恢复全比特深度视频。我们的方法在传感器端带来了内存效率、功耗节省与数据吞吐量降低方面的显著收益，使其非常适用于具有超低功耗需求或高速视频采集的成像系统。我们在由模拟 1 比特测量进行视频恢复的任务上验证了该框架。我们的工作重新定义了相机流水线，有望为基于低功耗传感器硬件的千兆像素、千赫兹成像系统铺平道路。
英文原文摘要 1178 字符
We introduce a fundamentally new paradigm in video sensing, 1-bit computational video, that redefines the limits of imaging efficiency and performance. Instead of the conventional high-bit-depth capture, we show that one bit measurements captured by time-varying thresholding can be used to reconstruct full-bit-depth videos, eliminating the need for power-hungry, high-precision analog-to-digital conversion (ADC) at the sensor as well as reducing the energy consumption in data transmission. We propose thresholding strategies to effectively capture spatiotemporal dependencies in video streams. Despite the significant data compression at acquisition, we recover full-bit-depth videos with high fidelity through neural video reconstruction. Our method unlocks significant gains in memory efficiency, power savings, and data throughput reduction at the sensor, making it ideal for imaging systems with ultra-low-power requirements or high-speed video capture. We validate our framework on video recovery from simulated 1-bit measurements. Our work redefines the camera pipeline, potentially paving the way for gigapixel, kilohertz imaging systems on low-power sensor hardware.