Point Cloud Editing Techniques

Point Cloud Editing

• Editing point clouds can be done in three ways:
  • Using the point cloud editing tool
  • Using the clipping box
  • Using the image annotation tool
• When editing point clouds, it's essential to orient yourself to avoid deleting points behind the ones you want to delete
• The "Paulaen trick" involves selecting an area, creating a new point group, and then deleting the points
• Inverting the selection and then deleting the points is another method

Image Annotation Tool

• The image annotation tool is a complex tool that allows for precise editing of points
• It can be used to delete points that are not desired, such as noise on a roof
• The tool allows for annotating images, which can then be used to delete points
• The image annotation tool can also be used to delete points that are not part of the main structure

3D Textured Mesh

• A 3D textured mesh is how models and maps are visualized for clients
• Generating a 3D textured mesh can be done without reprocessing step two by unchecking the step two box
• The 3D textured mesh fills in gaps between points by drawing triangles between them
• Noise in the model can be reduced by using spherical points instead of squares

Video Animation

• Video animation can be used to visualize maps and models
• A storyboard is a sequence of shots for a movie or film, and video animation can be used to create a storyboard
• Video animations can be exported as Waypoint trajectories for use with drones
• To create a video animation, the properties tab should be closed to ensure the correct aspect ratio
• There are two ways to create a video animation: user recorded views and computer camera positions and orientations
• User recorded views allow for more cinematographic control
• Interpolation can be used to create smoother video footage

Exporting Video Animations

• To export a video animation, the render option should be used
• The video animation can be saved as an mp4 file

Point Cloud Editing

• Three ways to edit point clouds: using the point cloud editing tool, clipping box, or image annotation tool
• Importance of orienting oneself to avoid deleting points behind desired points
• "Paulaen trick": select an area, create a new point group, and delete points
• Alternative method: invert selection and delete points

Image Annotation Tool

• A complex tool for precise point cloud editing
• Used to delete unwanted points, such as noise on a roof
• Allows annotating images to delete points
• Enables deletion of points not part of the main structure

3D Textured Mesh

• 3D visualization method for clients
• Generating a 3D textured mesh: uncheck step two box to skip reprocessing
• Mesh fills gaps between points by drawing triangles
• Noise reduction: use spherical points instead of squares

Video Animation

• Visualization method for maps and models
• Uses storyboarding for shot sequencing
• Exportable as Waypoint trajectories for drones
• To create: close properties tab for correct aspect ratio
• Two creation methods: user recorded views and computer camera positions
• User recorded views offer more cinematographic control
• Interpolation creates smoother footage

Exporting Video Animations

• Use render option to export video animation
• Save as an mp4 file

Merging Projects and Creating Manual Tie Points

• Ensure the same output coordinate system in both projects before merging, and convert units if necessary (e.g., from meters to feet).
• Create manual tie points (MTPs) with the same name, syntax, and location in both projects, aiming for 4-5 MTPs in each project.

Creating Manual Tie Points

• Select non-shadowed areas with easily discernable features (e.g., corners or flat surfaces) for MTPs.
• Avoid creating MTPs on elevated points (e.g., tables, gutters) and in shadowed areas.
• Use a consistent naming convention for MTPs, either keeping the original name or using a descriptive name (e.g., "northwest corner roof").
• Mark MTPs in multiple images (at least 6, but 8-10 recommended).

Processing and Reoptimizing

• Reoptimize the project after creating MTPs to ensure point cloud accuracy.
• Ensure the same processing template is used for both projects when reoptimizing.
• Limit reoptimization attempts to three to prevent losing the point cloud.

Merging Projects

• Create a new project and select "Project > Merge from existing projects" to merge two projects.
• Add the projects to be merged, sorted by date modified.
• Treat tie points with the same labels as identical.
• Process the point cloud after merging and review the quality report.

Point Cloud and Quality Report

• Disconnected point clouds may indicate incorrect MTPs.
• Review the quality report to ensure a successful merge.
• Delete unnecessary features and adjust the point cloud for better visualization if necessary.

Exercise 4: Mapping a Drone Field

• The exercise is a simplified image set designed to help understand the workflow of a complete project, with the goal of accurately mapping the drone field using Ground Control Points (GCPs).

Steps to Complete the Exercise

• Start a new project and import images into the software.
• Change the output coordinate system to the required format.
• Run step one of the workflow process.
• Import and mark Ground Control Points (GCPs) in the software.
• Re-optimize the project and then run step two of the workflow process.
• Produce a point cloud in LAS format as the final output.

Notes on the Challenge

• The exercise has a 30-minute time limit to complete the challenge.
• The goal of the challenge is to achieve the lowest RMS error possible.
• The winner of the challenge will receive a prize, specifically a landing pad.
• In the event of a tie, an additional exercise will be conducted to determine the winner.

Importance of GCPs

• Ground Control Points (GCPs) are easier to mark from drone images compared to white and black GCPs or arrow points.
• The GPS unit is highly accurate, making it easy to mark GCPs.

Final Instructions

• When the challenge is complete, raise your hand to indicate completion.
• Provide the quality report from step two of the workflow process.
• Do not share the quality report from step one of the process.

Exercise 5: Mapping and Measuring

• The 2nd Annual Drone Flying Event is used as a case study to demonstrate mapping and measuring techniques.
• Mr. Bill English from NTSB mapped the scene with a high-density Point Cloud after a car accident.

Taking Measurements

• Zoom in closely to ensure accurate measurements.
• Use scale constraints and tape measures to mark the same side of each object.
• Take measurements on flat objects, avoiding shadowed areas.

Creating Manual Tie Points (MTPs)

• Create an MTP by clicking on the point cloud and marking 8-10 images in a circular format.
• Apply changes to create a new point.

Drawing Polylines

• Draw a polyline by hovering over the new polyline button and clicking on the first MTP.
• Move to the second MTP and click to create a polyline.
• The polyline displays the terrain 3D length and projected 2D length.

Exporting Polylines as a Shape File

• Export polylines by going to the objects menu, selecting polylines, and right-clicking to export all polylines.
• Choose the ESRI Shapefile option to export the polylines.

Best Practices

• Choose distinct colors for marking points to avoid color bleed.
• Zoom in closely to ensure accurate marking.
• Use the minimum physical size under display properties to make MTPs larger and easier to work with.
• Mark points on the ground, avoiding elevated objects for accurate measurements.

Real-World Applications

• Accurate measurements are crucial in accident investigations, construction, and other industries.
• Exporting polylines as a shape file enables easy integration with other software, such as ArcGIS.

Visualizing Maps and Models

• The Randy Travis estate in Santa Fe is an example of an interactive model that can be visualized using drones and Matterport cameras.
• The model allows navigation of the interior and exterior of the property, including specific areas like the living room, bathroom, and balcony.
• Matterport cameras can scan interior areas, and the Insta 1X camera can be used via the Matterport app to scan interiors at a lower cost.

Uploading to SketchFab

• To upload a 3D map to SketchFab, select the FBX file in the "Upload" section.
• Uploaded models can be customized with keywords, descriptions, and backlinks to the website.
• Models can be shared with others, and descriptions can provide additional information about the service provided.

Processing and Cleaning Up Point Cloud

• Processing and cleaning up point cloud data is crucial for creating a good 3D map.
• Processing options can be adjusted to improve map quality, including the number of matches, minimum number of matches, and octree level.
• Cleaning up point cloud data is a time-consuming process, but it is essential for creating high-quality 3D maps.

Creating a 3D Background

• A 3D background can be added to the model to enhance visualization.
• Backgrounds can be created using an image or a 3D model.
• Post-processing filters can be adjusted to improve color and quality of the model.

Measuring Volumes

• Volumetric measurements can be used to measure the volume of a pile or a container.
• The base layer is critical for measuring volumes, and it can be triangulated or fitted to a plane.
• Volume settings can be adjusted to change the base surface, with options including triangulated, fit plane, align with the average altitude, align with the lowest point, or align with the highest point.

Delivering Results

• Results can be delivered to clients in a PDF format, including the ortho mosaic, area size, ground sampling distance, and quality report.
• The quality report can include the RMS error, which measures the accuracy of the model.
• The report can be used to showcase the quality of the work and provide a detailed analysis of the results.

Advanced Mapping Resources

• Additional resources, including videos and tutorials, are available for advanced mapping.
• These resources can be used to improve skills and knowledge of drone mapping and 3D modeling.
• Resources can also be used to provide a more detailed analysis of the results and improve the quality of the work.

Practice and Self-Awareness

• Practice is essential for improving skills and knowledge of drone mapping and 3D modeling.
• Self-awareness and hard work are also important for achieving success in the field.
• The formula to success is to practice, practice, practice, and to be aware of one's strengths and weaknesses.

Learn about different methods for editing point clouds, including using specific tools, clipping boxes, and image annotation. Discover tips and tricks for effective point cloud editing.