Code Modules:
- ImageProcessBPFunctionLibrary [Runtime]
Number of C++ Classes: 2
Network Replicated: No
Supported Development Platforms: (“Windows”)
Supported Target Build Platforms: ("Win64", "Win32")
Doc: Link
Example: Link
代码模块:
- ImageProcessBPFunctionLibrary[运行时]
C++类数:2
网络复制:没有
支持的开发平台:(”Windows”)
支持的目标构建平台:(”Win64″,”Win32″)
医生: 连结
例子:: 连结
Process the image -color mapping, image contour recognition, mosaic, frosted glass, oil painting, sketch, relief, lens, blur, face detection, cat face detection, license plate detection, etc.
Open the Content Browser window, in the upper right corner of the Content Browser, click the Settings button, This will open a menu where you can adjust various settings for the current instance of the Content Browser, Under the View category, check the Show Plugin Content option, Find the ImageProcess->Content->ImageProcess directory, There are sample levels and related assets inside.
Change log : (2023/06)
Support 5.2 version
5.1-5.2:
Added RenderTargetToTexture2D function.
Add MediaTextureToTexture2D function.
Note that if the conversion is performed every frame or some calculation-intensive functions will cause stuttering.
In UE5, you need to add plugin content in Additional Asset Directories to Cook in the project settings. Otherwise, it will crash when used after packaging. And when loading pictures, try to use PNG pictures, JPG still has problems that have not been solved.
Including some common functions (get all the file names in the path, load and save the picture file, crop the picture, superimpose the picture, reset the picture size, etc.)
- Most functions have the original function and the function P and the function TMS.
- FunctionP==>>The input image can be the disk path where the image is located.
- FunctionTMS==>>The input image can be the FTAndM structure returned by other functions. (Mainly used to accept Mat).
- The face detection function will have different accuracy rates based on different cascaded classifiers.
- Most area parameters are FBox, from Min coordinate point to Max coordinate point. For example, if it is Min (100, 100)-Max (100, 100), then the area of this area is equal to 0 instead of taking the width 100 and height 100 from Min (100, 100).
- Numerical parameters generally have default values. If the value is modified to a wrong value and the wrong result is obtained, the value can be restored to the default value.
- When setting the texture properties, you also need to set the value of MipGenSettings to NoMipmaps.
- GetPathFiles–Get all file names in the disk path
- LoadFileToTexture2D–Add image files to Texture2D through the disk file path
- SaveTexture2DToFile–Save Texture2D to the disk file of the corresponding path, and determine the file type according to the file name
- ResizeTexture2D–Reset the size of Texture2D
- CutOutTexture2D–Cut out a part of the area from Texture2D
- OverlayTexture2D–Overlay another Texture2D to the specified area for Texture2D
- MakeTAndM–Create a TAndM
- FindContours–Identify the contour in the image and return the point data of the contour
- ApplyColorMap–Apply a color map to Texture2D
- Mosaic–Apply a mosaic filter to the specified area for Texture2D
- FrostedGlass–Apply a Frosted glass filter to the specified area for Texture2D
- OilPainting–Apply a Oil Painting filter to the specified area for Texture2D
- Sketch–Convert Texture2D to sketch filter
- ReliefSculpture–Convert Texture2D to Relief Sculpture filter
- Lens–Convert Texture2D to Lens filter
- Blurred–Convert Texture2D to Blurred filter
- FaceDetection–According to the use of different CascadeClassifier to carry out different target detection.
处理图像-颜色映射,图像轮廓识别,马赛克,磨砂玻璃,油画,素描,浮雕,镜头,模糊,人脸检测,猫脸检测,车牌检测等。
打开Content Browser窗口,在Content Browser的右上角,点击Settings按钮,这将打开一个菜单,您可以在其中调整Content Browser当前实例的各种设置,在View类别下,选中Show Plugin Content选项,找到ImageProcess->Content->ImageProcess目录,
更改日志:(2023/06)
支持5.2版本
5.1-5.2:
增加了RenderTargetToTexture2D函数。
添加MediaTextureToTexture2D函数。
请注意,如果每帧执行转换或某些计算密集型功能会导致口吃。
在UE5中,您需要在其他资源目录中添加插件内容,以便在项目设置中进行烹饪。 否则,包装后使用时会崩溃。 而在加载图片的时候,尽量使用PNG图片,JPG还是有问题没有解决。
包括一些常用的功能(获取路径中所有的文件名,加载保存图片文件,裁剪图片,叠加图片,重置图片大小等。)
- 大多数函数具有原始函数以及函数P和函数TMS。
- FunctionP==>>输入映像可以是映像所在的磁盘路径。
- FunctionTMS==>>输入图像可以是其他函数返回的FTAndM结构。 (主要用于接受垫)。
- 人脸检测功能会根据不同的级联分类器而具有不同的准确率。
- 大多数区域参数都是FBox,从最小坐标点到最大坐标点。 例如,如果它是Min(100,100)-Max(100,100),那么该区域的面积等于0,而不是从Min(100,100)取宽度100和高度100。
- 数值参数通常具有默认值。 如果该值被修改为错误的值并且获得了错误的结果,则该值可以恢复为默认值。
- 设置纹理属性时,还需要将MipGenSettings的值设置为NoMipmaps。
- GetPathFiles–获取磁盘路径中的所有文件名
- LoadFileToTexture2D–通过磁盘文件路径向Texture2D添加图像文件
- SaveTexture2DToFile–将Texture2D保存到相应路径的磁盘文件中,根据文件名确定文件类型
- ResizeTexture2D–重置Texture2D的大小
- CutOutTexture2D–从Texture2D中切出一部分区域
- OverlayTexture2D–将另一个Texture2D叠加到Texture2D的指定区域
- MakeTAndM–创建一个TAndM
- FindContours–识别图像中的轮廓,返回轮廓的点数据
- ApplyColorMap–将颜色映射应用于Texture2D
- Mosaic–对Texture2D的指定区域应用一个mosaic滤镜
- FrostedGlass–在Texture2D的指定区域应用磨砂玻璃滤镜
- OilPainting–在Texture2D指定区域应用油画滤镜
- Sketch–将Texture2D转换为sketch filter
- ReliefSculpture–将Texture2D转换为Relief Sculpture filter
- 镜头–将Texture2D转换为镜头滤镜
- 模糊–将Texture2D转换为模糊滤镜
- FaceDetection–根据使用不同的CascadeClassifier进行不同的目标检测。
评论(0)