想要的效果就是将marker变成一个窗口，通过这个窗口可以看见里面的虚拟内容
做的东西有点类似于传送门，但与网上常见的AR传送门又有一点不同，网上常见的AR传送门其中“门”的位置通常是任意的，b站上有一个老哥的视频，可以看看AR传送门的效果。
【附源码】Unity+ARcore 实现传送门 小意思VR AR教程
我想做的这个“门”是和marker同样大小的，有点像官方的一个演示视频（官方的这个示例里从marker到虚拟物体的过渡做的很流畅）
还有之前看到的一个国外网友做的东西。
效果大概就是这个样子，最终做出的demo：
识别marker
ARCore组件中的Example下有一个名为增强图像（Augmented Image）的示例，展示了ARCore如何识别marker并在上面显示模型。初学记录一下做的过程和理解。
代码主要有两部分AugmentedImageExampleController和AugmentedImageVisualizer。AugmentedImageVisualizer部分是ARCore检测到marker图像之后，负责在其上显示模型的组件。就像HelloAR中显示模型一样。
能够发现在AugmentedVisualizer脚本上留有左下角、右下角、左上角和右上角这四个部分的模型等待绑定。
Prefab目录下的AugmentedImageVisualizer上绑定了同名的script，并为其绑定了对应的四个模型。 在prefab的Inspector中我们看到的如下。
prefab与实际使用时看到的不一样是因为在Visualizer的script中对4部分的位置都分别做了调整。
然后在AugmentedImageExampleController中绑定了上面的prefab
初见遇到的坑
这些坑太蠢了，笑笑跳过去吧。
1、换成自己的模型之后，在Visualizer脚本中试图修改模型的缩放和位置，但是没有效果，模型显示的时候还是出现在初始位置且是初始的缩放比例。
最初的示例Augmented Image中，当改变marker的大小时，模型的位置和大小也会发生改变，也就是说，从一开始模型和marker的相对大小就是固定的，而脚本中有修改模型位置的代码。我的问题就在于，这个修改位置的代码不起作用。
是因为我在Inspector上脚本处绑定物体不对。我绑定的是模型文件夹下的，而不是prefab内的模型？二者是不同的，相互独立的，虽然它们大小坐标等等是一样的，但是在场景中显示的是prefab内的模型，我们修改的也应当是prefab的组成部分，也就是说之前修改的统统没有使用。
对！就是这里。
2、关于localscale缩放大小的问题，是我蠢逼了，要么在script中设置，要么在Inspector中设置，不要两头同时做。被自己蠢到了，计算了一晚上找不出到底哪里错了，还搁这研究坐标变换呢....
坐标系
就算不熟悉ARCore、unity的坐标系，通过Vector3的forward、back等6个向量加上prefab中四个模型和Visualizer脚本中的代码，我们也可以推出来坐标系的样子。如下图：
unity中的世界坐标系是左手坐标系，用左手做出一个手枪枪毙的动作，将中指指向掌心方向，大拇指代表X轴，食指代表Y轴，中指代表Z轴。
6个单位向量的xyz值
Vector3.forward (0,0,1)
Vector3.left (-1,0,0)
Vector3.up (0,1,0)
back，right，down分别相反
着色器Shader
网上已经有很多这种“透明物体遮挡实体”的效果了，参考：
https://jingyan.baidu.com/article/9faa7231f98071473c28cb85.html
https://blog.csdn.net/u014361280/article/details/103690369
最近正在看《Unity Shader入门精要》这本，微信读书上就有，对入门算是很友好了。
为了实现透明且同时遮挡后面物体的效果，需要单独创建一个Shader，使用VS打开，修改其代码
Shader "Unlit/MaskShader"
{
SubShader
{
Tags { "Queue" = "Geometry-10"}
Lighting off
ZTest LEqual
ZWrite On
ColorMask 0
Pass{}
}
}
与传送门的不同
与传送门不同的是，要更改“门”的大小到和marker一致，这个很简单，因为ARCore能够估计marker的宽高并返回值。
public AugmentedImage Image;
可通过Image的ExtentX和ExtentZ属性来获得ARCore对增强图像尺寸的估计
C#中结构体struct的坑
C#中的结构体
localScale.x = 1.0f;
这样是不行的，要修改必须将localScale整个修改，可以搜索关键字“C#结构体”来了解一下这个问题。
localScale = new Vector3(x,y,z);
代码
Visualizer：
//-----------------------------------------------------------------------
// <copyright file="AugmentedImageVisualizer.cs" company="Google LLC">
//
// Copyright 2018 Google LLC. All Rights Reserved.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
// </copyright>
//-----------------------------------------------------------------------
namespace GoogleARCore.Examples.AugmentedImage
{
using System;
using System.Collections.Generic;
using System.Runtime.InteropServices;
using GoogleARCore;
using GoogleARCoreInternal;
using UnityEngine;
/// <summary>
/// Uses 4 frame corner objects to visualize an AugmentedImage.
/// </summary>
public class MyVisualizer : MonoBehaviour
{
/// <summary>
/// The AugmentedImage to visualize.
/// </summary>
//增强图像Image
public AugmentedImage Image;
//box变成marker的几倍
//必须为大于等于1的值，最好大于等于2
public float box_scale = 3.0f;
//box_depth为box的深度
//注意box要下降的深度应当为box_depth的一半，同时roof是不需要下降的
public float box_depth = 3.0f;
/// <summary>
/// A model for the lower left corner of the frame to place when an image is detected.
/// </summary>
///
public GameObject roof_left;
public GameObject roof_right;
public GameObject roof_up;
public GameObject roof_down;
public GameObject wall_forward;
public GameObject wall_back;
public GameObject wall_left;
public GameObject wall_right;
public GameObject wall_bottom;
public GameObject ball_1;
public GameObject ball_2;
public GameObject ball_3;
public GameObject ball_4;
public GameObject Mask_roof_left;
public GameObject Mask_roof_right;
public GameObject Mask_roof_up;
public GameObject Mask_roof_down;
public GameObject Mask_wall_forward;
public GameObject Mask_wall_back;
public GameObject Mask_wall_left;
public GameObject Mask_wall_right;
public GameObject Mask_wall_bottom;
/// <summary>
/// The Unity Update method.
/// </summary>
//unity的刷新函数update是从Monobehaviour类中继承来的
//当检测到增强图像时，这个update函数才会不断调用（因为这个函数的用处就是显示在增强图像上的模型）
public void Update()
{
// UnityEngine.Debug.Log("update MyVisualizer");
//1、当没有预先设置图像时
//2、图像检索的状态并不是搜索中
//将四个模型的Active都设置为False
if (Image == null || Image.TrackingState != TrackingState.Tracking)
{
roof_left.SetActive(false);
roof_right.SetActive(false);
roof_up.SetActive(false);
roof_down.SetActive(false);
wall_forward.SetActive(false);
wall_back.SetActive(false);
wall_left.SetActive(false);
wall_right.SetActive(false);
wall_bottom.SetActive(false);
ball_1.SetActive(false);
ball_2.SetActive(false);
ball_3.SetActive(false);
ball_4.SetActive(false);
Mask_wall_forward.SetActive(false);
Mask_wall_back.SetActive(false);
Mask_wall_left.SetActive(false);
Mask_wall_right.SetActive(false);
Mask_wall_bottom.SetActive(false);
Mask_roof_left.SetActive(false);
Mask_roof_right.SetActive(false);
Mask_roof_up.SetActive(false);
Mask_roof_down.SetActive(false);
return;
}
float roof_thickness = 0.0001f;
//估计图像（图像是2维的）物理宽度、高度的一半
float halfWidth = Image.ExtentX / 2;
float halfHeight = Image.ExtentZ / 2;
//mask缩放和移动时使用的偏移量
float offset_for_mask_scale = 0.005f;
float offset_for_mask_pos = 0.001f;//此处应和墙的厚度一致，不然会出现闪烁
//临时存储数据
float temp_for_mask_x;
float temp_for_mask_z;
Vector3 temp;
//ball
float min = Math.Min(halfHeight, halfWidth);
ball_1.transform.localScale =
new Vector3(0.01f,0.01f,0.01f);
temp = ball_2.transform.localScale;
temp.x = min;
temp.y = min ;
temp.z = min ;
ball_2.transform.localScale =
new Vector3(temp.x, temp.y, temp.z);
temp = ball_3.transform.localScale;
temp.x = min ;
temp.y = min ;
temp.z = min ;
ball_3.transform.localScale =
new Vector3(temp.x, temp.y, temp.z);
temp = ball_4.transform.localScale;
temp.x = min ;
temp.y = min ;
temp.z = min ;
ball_4.transform.localScale =
new Vector3(temp.x,temp.y,temp.z);
//roof_left和right的缩放
temp = roof_left.transform.localScale;
temp.x = halfWidth * (box_scale - 1);
temp.z = halfHeight * 2 * box_scale;
temp_for_mask_x = temp.x + offset_for_mask_scale;
temp_for_mask_z = temp.z + offset_for_mask_scale;
roof_left.transform.localScale = new Vector3(temp.x, roof_thickness, temp.z);
roof_right.transform.localScale = new Vector3(temp.x, roof_thickness, temp.z);
Mask_roof_left.transform.localScale = new Vector3(temp_for_mask_x, roof_thickness, temp_for_mask_z);
Mask_roof_right.transform.localScale = new Vector3(temp_for_mask_x, roof_thickness, temp_for_mask_z);
//roof up和down的缩放
temp = roof_up.transform.localScale;
temp.x = halfWidth * 2;
temp.z = halfHeight *(box_scale-1);
temp_for_mask_x = temp.x + offset_for_mask_scale;
temp_for_mask_z = temp.z + offset_for_mask_scale;
roof_up.transform.localScale = new Vector3(temp.x, roof_thickness, temp.z);
roof_down.transform.localScale = new Vector3(temp.x, roof_thickness, temp.z);
Mask_roof_up.transform.localScale = new Vector3(temp_for_mask_x, roof_thickness, temp_for_mask_z);
Mask_roof_down.transform.localScale = new Vector3(temp_for_mask_x, roof_thickness, temp_for_mask_z);
//forward和back的缩放
temp = wall_forward.transform.localScale;
temp.x = halfWidth * 2* box_scale;
temp_for_mask_x = temp.x + offset_for_mask_scale;
temp.y = box_depth;
wall_forward.transform.localScale = new Vector3(temp.x, temp.y, temp.z);
wall_back.transform.localScale = new Vector3(temp.x, temp.y, temp.z);
Mask_wall_forward.transform.localScale = new Vector3(temp_for_mask_x, temp.y, temp.z);
Mask_wall_back.transform.localScale = new Vector3(temp_for_mask_x, temp.y, temp.z);
//left和right的缩放
temp = wall_left.transform.localScale;
temp.z = halfHeight * 2* box_scale;
temp_for_mask_z = temp.z + offset_for_mask_scale;
temp.y = box_depth;
wall_left.transform.localScale = new Vector3(temp.x, temp.y, temp.z);
wall_right.transform.localScale = new Vector3(temp.x, temp.y, temp.z);
Mask_wall_left.transform.localScale = new Vector3(temp.x, temp.y, temp_for_mask_z);
Mask_wall_right.transform.localScale = new Vector3(temp.x, temp.y, temp_for_mask_z);
//bottom的缩放
temp = wall_bottom.transform.localScale;
temp.x = halfWidth * 2* box_scale;
temp.z = halfHeight * 2* box_scale;
temp.y = box_depth;
wall_bottom.transform.localScale = new Vector3(temp.x, roof_thickness, temp.z);
Mask_wall_bottom.transform.localScale = new Vector3(temp_for_mask_x, roof_thickness, temp_for_mask_z);
ball_1.transform.localPosition =
(box_depth / 2 - ball_2.transform.localScale.y / 2) * Vector3.down;
ball_2.transform.localPosition =
-halfWidth*box_depth*5/2*Vector3.left + (box_depth/2 - ball_2.transform.localScale.y/2)*Vector3.down + halfHeight*box_depth*5/2*Vector3.forward;
ball_3.transform.localPosition =
halfWidth*box_depth*5/2* Vector3.left + (box_depth / 2 - ball_2.transform.localScale.y / 2) * Vector3.down + halfHeight*box_depth*5/2 * Vector3.back;
ball_4.transform.localPosition =
(box_depth / 2 - ball_2.transform.localScale.y / 2) * Vector3.down;
// (halfHeight * (box_scale + 1) / 2 + offset_for_mask_pos) * Vector3.forward;
//修改roof的位置
roof_up.transform.localPosition =
(halfHeight * (box_scale + 1) / 2 + offset_for_mask_pos) * Vector3.forward;
roof_down.transform.localPosition =
(halfHeight * (box_scale + 1) / 2 + offset_for_mask_pos) * Vector3.back;
roof_left.transform.localPosition =
(halfWidth * (box_scale + 1)/2 + offset_for_mask_pos)*Vector3.left;
roof_right.transform.localPosition =
(halfWidth * (box_scale + 1)/2 + offset_for_mask_pos) * Vector3.right;
//修改roof mask的位置
Mask_roof_up.transform.localPosition =
(offset_for_mask_pos * Vector3.up) + halfHeight * (box_scale + 1) / 2 * Vector3.forward;
Mask_roof_down.transform.localPosition =
(offset_for_mask_pos * Vector3.up) + halfHeight * (box_scale + 1) / 2 * Vector3.back;
Mask_roof_left.transform.localPosition =
(offset_for_mask_pos * Vector3.up) + halfWidth * (box_scale + 1) / 2 * Vector3.left;
Mask_roof_right.transform.localPosition =
(offset_for_mask_pos * Vector3.up) + halfWidth * (box_scale + 1) / 2 * Vector3.right;
//修改位置，让这个box“下沉”，陷入到marker里面。
wall_forward.transform.localPosition =
(box_depth/2 * Vector3.down) + (box_scale*halfHeight * Vector3.forward);
wall_back.transform.localPosition =
(box_depth/2 * Vector3.down) + (box_scale * halfHeight * Vector3.back);
wall_left.transform.localPosition =
(box_depth/2 * Vector3.down) + (box_scale * halfWidth * Vector3.left);
wall_right.transform.localPosition =
(box_depth/2 * Vector3.down) + (box_scale * halfWidth * Vector3.right);
wall_bottom.transform.localPosition = box_depth/2 * Vector3.down;
//修改mask的位置，保证能够覆盖墙面的同时，还要不与墙面重叠（否则会有闪烁）
Mask_wall_forward.transform.localPosition =
(box_depth / 2 * Vector3.down) + ((box_scale * halfHeight + offset_for_mask_pos) * Vector3.forward);
Mask_wall_back.transform.localPosition =
(box_depth / 2 * Vector3.down) + ((box_scale * halfHeight + offset_for_mask_pos) * Vector3.back);
Mask_wall_left.transform.localPosition =
(box_depth / 2 * Vector3.down) + ((box_scale * halfWidth + offset_for_mask_pos) * Vector3.left);
Mask_wall_right.transform.localPosition =
(box_depth / 2 * Vector3.down) + ((box_scale * halfWidth + offset_for_mask_pos) * Vector3.right);
Mask_wall_bottom.transform.localPosition = (box_depth+ offset_for_mask_pos) * Vector3.down;
roof_left.SetActive(true);
roof_right.SetActive(true);
roof_up.SetActive(true);
roof_down.SetActive(true);
Mask_roof_left.SetActive(true);
Mask_roof_right.SetActive(true);
Mask_roof_up.SetActive(true);
Mask_roof_down.SetActive(true);
wall_forward.SetActive(true);
wall_back.SetActive(true);
wall_left.SetActive(true);
wall_right.SetActive(true);
wall_bottom.SetActive(true);
ball_1.SetActive(true);
ball_2.SetActive(true);
ball_3.SetActive(true);
ball_4.SetActive(true);
Mask_wall_forward.SetActive(true);
Mask_wall_back.SetActive(true);
Mask_wall_left.SetActive(true);
Mask_wall_right.SetActive(true);
Mask_wall_bottom.SetActive(true);
}
}
}
Controller：
//-----------------------------------------------------------------------
// <copyright file="AugmentedImageExampleController.cs" company="Google LLC">
//
// Copyright 2018 Google LLC. All Rights Reserved.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
// </copyright>
//-----------------------------------------------------------------------
namespace GoogleARCore.Examples.AugmentedImage
{
using System.Collections.Generic;
using System.Runtime.InteropServices;
using GoogleARCore;
using UnityEngine;
using UnityEngine.UI;
/// <summary>
/// Controller for AugmentedImage example.
/// </summary>
/// <remarks>
/// In this sample, we assume all images are static or moving slowly with
/// a large occupation of the screen. If the target is actively moving,
/// we recommend to check <see cref="AugmentedImage.TrackingMethod"/> and
/// render only when the tracking method equals to
/// <see cref="AugmentedImageTrackingMethod"/>.<c>FullTracking</c>.
/// See details in <a href="https://developers.google.com/ar/develop/c/augmented-images/">
/// Recognize and Augment Images</a>
/// </remarks>
/// 一个增强图像示例类
public class MyController : MonoBehaviour
{
/// <summary>
/// A prefab for visualizing an AugmentedImage.
/// </summary>
// 隔壁的MyVisualizer类创建一个预制体AugmentedImageVisualizerPrefab
public MyVisualizer AugmentedImageVisualizerPrefab;
/// <summary>
/// The overlay containing the fit to scan user guide.
/// </summary>
// 用来指引用户的覆盖层
public GameObject FitToScanOverlay;
//创建一个私有的字典对象，int为索引，图像为内容
//并对每个都初始化 每一个对应一张图片
private Dictionary<int, MyVisualizer> m_Visualizers
= new Dictionary<int, MyVisualizer>();
private List<AugmentedImage> m_TempAugmentedImages = new List<AugmentedImage>();
/// <summary>
/// The Unity Awake() method.
/// </summary>
public void Awake()
{
UnityEngine.Debug.Log("example awake");
// Enable ARCore to target 60fps camera capture frame rate on supported devices.
// Note, Application.targetFrameRate is ignored when QualitySettings.vSyncCount != 0.
//垂直同步不为0时，目标帧率是无效的
Application.targetFrameRate = 60;
}
/// <summary>
/// The Unity Update method.
/// </summary>
public void Update()
{
//UnityEngine.Debug.Log("update example");
// Exit the app when the 'back' button is pressed.
//检查是否退出
if (Input.GetKey(KeyCode.Escape))
{
Application.Quit();
}
// Only allow the screen to sleep when not tracking.
//当不追踪图像时允许屏幕睡眠
//一共有几种Status？
if (Session.Status != SessionStatus.Tracking)
{
Screen.sleepTimeout = SleepTimeout.SystemSetting;
}
else
{
Screen.sleepTimeout = SleepTimeout.NeverSleep;
}
// Get updated augmented images for this frame.
//GetTrackable函数返回可追踪对象，存储到列表m_TempAugmentedImage
//类型为增强图像AugmentedImage，还可以是其他的类型（比如增强面部，检测的平面，特征点）
//使用TrackableQueryFilter.Updated对返回值进行过滤
Session.GetTrackables<AugmentedImage>(
m_TempAugmentedImages, TrackableQueryFilter.Updated);
// Create visualizers and anchors for updated augmented images that are tracking and do
//为每一个正在追踪的增强图像创建可视化工具和锚点
// not previously have a visualizer. Remove visualizers for stopped images.
//
foreach (var image in m_TempAugmentedImages)
{
MyVisualizer visualizer = null;
m_Visualizers.TryGetValue(image.DatabaseIndex, out visualizer);
//TrackingState是枚举类型吗
if (image.TrackingState == TrackingState.Tracking && visualizer == null)
{
// Create an anchor to ensure that ARCore keeps tracking this augmented image.
//创建锚点来使ARCore持续追踪增强图像marker
//在图像的中心创建锚点
Anchor anchor = image.CreateAnchor(image.CenterPose);
//强制类型转换为marker可视化对象
//使用之前创建的预制体+锚点的方向
//Instantiate实例化，这是unity的函数，将一个对象及其子物体完全复制，生成一个新的对象（包括坐标），第一个参数是被复制的对象，第二个参数就是为其赋值新的坐标
//将预制体进行复制，坐标为每个图形锚点的坐标
visualizer = (MyVisualizer)Instantiate(
AugmentedImageVisualizerPrefab, anchor.transform);
//将增强图像传给可视化组件
visualizer.Image = image;
//将可视化组件添加到 字典 m_Visualizers中存储起来
m_Visualizers.Add(image.DatabaseIndex, visualizer);
}
else if (image.TrackingState == TrackingState.Stopped && visualizer != null)
{
//如果不再追踪增强图像，则将其从字典中移除。并且释放对象
m_Visualizers.Remove(image.DatabaseIndex);
GameObject.Destroy(visualizer.gameObject);
}
}
// Show the fit-to-scan overlay if there are no images that are Tracking.
foreach (var visualizer in m_Visualizers.Values)
{
if (visualizer.Image.TrackingState == TrackingState.Tracking)
{
FitToScanOverlay.SetActive(false);
return;
}
}
FitToScanOverlay.SetActive(true);
}
}
}
最后
感觉用pad来显示marker，因为有屏幕反光的原因，会有一定的抖动。改天把marker打印出来试试抖动会不会小一些。打印出的marker是黑白的也没有关系。
检测仅基于高对比度的点，所以彩色和黑白图像都会被检测到，无论使用彩色还是黑白参考图像。
很简单的东西，因为自己太蠢踩了很多坑，其中大多是因为不熟悉unity和C#绕的弯路，感觉可以做点有意思的小玩意。
本文地址：https://blog.csdn.net/Vikanill/article/details/106955314
