This article goes step by step through the process of creating a new project in Pix4Dinspect.
A new project is a cloud-based project that belongs to an asset and allows generating Intelligent Digital twin, visualize them in 2D and 3D, inspect the reconstructed models and annotate images for collaboration.
IN THIS ARTICLE
How to create a project
Inputs selection
Project creation
Options
How to create a New Project
1. On the left of the Dashboard, next to the Assets viewer, click and select New Project.
2. The window Create New Project appears.
Inputs selection
Inputs can be added via drag and drop or selected from the disk.
Drag and drop input files
- Drag and drop the input files on the screen.
- (Optional) More input files can be added if needed.
- Click Next.
Select from Disk
- Click Select from Disk.
- (Optional) More input files can be added if needed.
- Click Next.
After the input images are selected, the below checks are carried out during the third step. The user will be redirected to the second step if the requirements are not met.
- For all Asset types: At least 5 images and/or 1 video
- Maximum 4,000 images
- Georeferenced images
- Images with compatible (XMP) camera orientation
Project creation
After you have selected your inputs, in the Project creation tab, you need to choose between two options: Existing asset or New asset.
Existing asset
- Select your asset from the drop-down list, or click an existing asset from the interactive map.
- (Optional) Give it a name for the Project.
- Select a Date.
- Click Next
New asset
- Type a Location (coordinates), or click a location from the interactive map.
- Give a Name for the asset.
- Select an Asset Type from the drop-down list.
- (Optional) Give a short description of the asset.
- (Optional) Give a name for the Project.
- Select a Date.
- Click Next.
Options
Different options are available to define the intelligent digital twin and to activate the artificial intelligence algorithms for antenna detection.
- Digital twin: select this option to generate all the following intelligent digital twin (depending on the selected asset you might get different outputs):
- 3D Model: The 3D model provides awareness for the image navigation and enables to perform any 3D measurement. The generation of the 3D model is mandatory for the execution of any Artificial Intelligence plugin.
- Textured 3D Model: A textured 3D model is based on faces and textures, contrary to the 3D model, which is a point cloud.
- Image map: The Image map is a 2D map where each point contains X, Y, and color information. It has a uniform scale and can be used for 2D measurements (distance, surface).
- Elevation map: It is a DSM, a 2.5 D model of the mapped area. For each (X,Y) position, the DSM has only 1 Z value (the altitude of the highest point for this (X,Y) position). This is why it is considered to be a 2.5D model.
- Area of interest: defines the radius of the circle outside of which the 3D Model will be cut. The center of the circular area in the center of the cell tower, which is estimated through a density structure analysis (depending on the selected asset you might get different options):
- Auto: crops the 3D model using the automatic image geolocation
- Don't Crop: recommended for a project mapping the vertical wall
- Custom: crops the 3D model using the custom radius. This option is only available for the Cell tower assets.
- Artificial Intelligence plugins:
- Tower Inventory: Automatically extract the inventory list of panel antennas, microwave antennas, and ancillary boxes. Also, the inventory information of angles, dimensions, and positions.
- Rooftop segmentation: Automatically detect and measure usable rooftop faces.
- Rust detection: Automatically detect rust-affected areas.
- Custom AI plugin: We can apply custom AI algorithms to your data and improve your intelligent digital twin. Contact us for more information.
The user can navigate between the different steps by clicking Previous and Next.
To start generating the outputs, click Upload.