捕获速度

3D应用中一个重要因素是速度。为了估计应用的运行速度，我们需要知道每个组件需要多少时间。有两个重要的测量项目可以估计 Zivid 需要多少时间：

捕获函数返回时间

这是执行 capture API 所需的时间。当 API 返回时，可以安全地阻挡相机的 FOV 或移动相机。在 Zivid Studio 中，CaptureFunctionReturnTime 显示为Acquisition Time（采集时间）。

捕获时间

这是从调用 capture API 到点云在 GPU 上可用所需的时间。它不包括将数据从 GPU 内存复制到系统内存。

关于捕捉过程的详细介绍，请查看 点云捕获过程 。

Zivid 对不同硬件上的所有预设设置进行了基准测试，无论是 高端或者低端 硬件配置。这些基准会随着时间的推移而变化，在大多数情况下会随着我们转向更新的 SDK 版本而有所改善。对于相关数值，请选择与您的 SDK 版本相对应的本文档版本。

您可以在 Zivid Studio 中以及通过 SDK 查看您的采集和捕获时间。

C++

source

const auto frame = camera.capture2D3D(settings);
const auto frameInfo = frame.info();
std::cout << "Acquisition time:" << std::endl;
std::cout << std::chrono::duration_cast<std::chrono::milliseconds>(
                 frameInfo.metrics().acquisitionTime().value())
                 .count()
          << " ms" << std::endl;
std::cout << "Capture time:" << std::endl;
std::cout
    << std::chrono::duration_cast<std::chrono::milliseconds>(frameInfo.metrics().captureTime().value()).count()
    << " ms" << std::endl;

C#

source

var frame = camera.Capture2D3D(settings);
var frameInfo = frame.Info;
Console.WriteLine("Acquisition time:");
Console.WriteLine(frameInfo.Metrics.AcquisitionTime.Milliseconds + " ms");
Console.WriteLine("Capture time:");
Console.WriteLine(frameInfo.Metrics.CaptureTime.Milliseconds + " ms");

Python

source

frame = camera.capture_2d_3d(settings)
frame_info = frame.info
print("Acquisition time:")
print(f"{frame_info.metrics.acquisition_time.total_seconds() * 1000:.0f} ms")
print("Capture time:")
print(f"{frame_info.metrics.capture_time.total_seconds() * 1000:.0f} ms")

您还可以运行我们的 C++ 示例 ZividBenchmark.cpp 来对您的硬件进行基准测试。

Cameras

Configurations

Low-end Intel High-end NVIDIA High-end Intel Low-end Jetson NVIDIA High-end Jetson NVIDIA

Details about configurations

Display Options

Capture CaptureFunctionReturn

Standard Deviation

Capture 2D + 3D (hide)

Manufacturing Settings Parcels Settings Consumer Goods Settings Inspection Settings

hidden

hidden

	2+R
	High-end NVIDIA
	2D	3D
Manufacturing Settings
Manufacturing Diffuse	30 ms (±0)155 ms (±1)180 ms (±1)
Manufacturing Semi-Specular	30 ms (±0)740 ms (±1)765 ms (±1)
Manufacturing Specular	47 ms (±0)960 ms (±0)1000 ms (±1)
Manufacturing Small Features	42 ms (±2)890 ms (±1)910 ms (±1)

3D

(hide)

Manufacturing Settings Parcels Settings Consumer Goods Settings Inspection Settings

hidden

hidden

	2+R
	High-end NVIDIA
Manufacturing Settings
Manufacturing Diffuse	175 ms (±2)
Manufacturing Semi-Specular	760 ms (±5)
Manufacturing Specular	985 ms (±17)
Manufacturing Small Features	905 ms (±5)

2D

(hide)

Exposure Time Settings 2D HighRes Settings 2D LowRes Settings 2D

hidden

hidden

ET=ExposureTime	2+R
	ET	High-end NVIDIA
Exposure Time Settings 2D
Exposure Time 5 ms	5 ms	40 ms (±2)
Exposure Time 10 ms	10 ms	60 ms (±2)
Exposure Time 15 ms	15 ms	80 ms (±3)
Exposure Time 20 ms	20 ms	100 ms (±3)
Exposure Time 25 ms	0 ms	0 ms (±0)
Exposure Time 30 ms	0 ms	0 ms (±0)
Exposure Time 40 ms	0 ms	0 ms (±0)
Exposure Time 50 ms	0 ms	0 ms (±0)
Exposure Time 75 ms	0 ms	0 ms (±0)
Exposure Time 100 ms	0 ms	0 ms (±0)

Capture 2D + 3D (hide)

Capture 2D and then 3D

Capture 3D and then 2D

Capture 3D including 2D

hidden

hidden

	2+R
	High-end NVIDIA
	2D	3D
Capture 2D and then 3D
Projector On for 2D	47 ms290 ms335 ms
Capture 3D and then 2D
Capture 3D including 2D

Filters

(hide)

hidden

hidden

	2+R
	High-end NVIDIA
Filters
ContrastDistortionCorrection	2 ms (±0)
ContrastDistortionRemoval	0.5 ms (±1)
Noise	0 ms (±0)
Outlier	0.2 ms (±1)
Cluster	2.8 ms (±1)
ReflectionGlobal	6 ms (±1)
Smoothing	0.7 ms (±0)
RegionOfInterest-Box	0.3 ms (±1)
RegionOfInterest-Depth	0.2 ms (±1)