Point Cloud Cleanup

Increase NavVis IVION adoption and engagement in the AEC market, by allowing users to rely on IVION to clean their capture data from unwanted information and reduce 3rd party software spending.

Lidar SLAM capturing reflections causes issues due to inaccurate data; it is a know problem that reduces trust in the information scanned. Currently customers rely on third party tools to solve this.

The goal is to allow current customers to do these tasks in IVION and engage prospects who are being pushed away due to the lack of this specific feature.

ROLE

Senior UX/UI designer

Strategy, User Research, Competitor analysis, stakeholder and project management,  Wireframing, UI/UX design, Usability testing, Customer interviews, workshop facilitation, KPI definition and Data analytics,

SLAM laser scan indoor reflection points caused by reflective surface of the window from outdoor scans. The trees from outside field appear inside

Problem: Reflections in ‘reality capturing’, don’t represent ‘reality’

Our current data capture process leads to inclusion of excessive and unnecessary points, resulting into inaccurate representation of reality and added workflow complexity. Additionally, the necessity to transfer data from NavVis IVION to other tools for point cloud cleanup results in a loss of connection to panoramas.

What do I mean when I say unnecessary points?
I’m referring specifically to reflection points, these artifacts appear when scanning reflective surfaces such as windows and mirrors. These do not affect the panoramic images captured with the device.

Several customers require the data captured to be transformed into BIM or mesh data. The inhability to remove these points in IVION forces the users to transfer the data to 3rd party softwares, but in doing so they can’t bring the data back into IVION without losing the connection to the panoramic images.

Prioritising the insights and use cases comings from CSMs and customers, we focused primarily on the the user story:

As a project admin, I want to remove the reflection point from my scanned data, so that my data correctly reflects reality and I can either deliver more accurate data to my customers or export cleaner data to be converted into 3D mesh.

Congression center hallway panoramic image, this photo represents the reality as it doesn’t show any artifacts like point cloud scan.

How to solve the problem: What are customers doing now, and what do they do with the data.

I was tasked to solve this problem, but to find the right solution I wanted to understand how is the problem solved today.

I started conducting competitors analysis and user research.

With the help of CSMs I was able to get information on the preferred competitor’s tools our customers use to solve the pain point: tools such as Faro scene; Leica cyclone; Cloud Compare; and more. I got an understanding what worked well and what could still be improved.

For the user research thanks to customer facing teams and our user panel, I conducted interviews directly with our customer’s data editors to find out where I could find solution opportunities in their workflow.

What I learned:

Using different softwares outside of IVION to cleanup their data leads to increased time on task (and increased software costs), as well as the loss of panoramic images connection when bringing the data back into NavVis IVION.

The best solution: build our internal cleanup tool to increase NavVis IVION engagement and adoption, helping current customer save time and money, as well as using this as a sellling point for acquisition.

Competitors' logos
Competitor's analysis table

competitor’s and 3rd party tools analysis table to understand what our competitors are offering and what our customer’s are using to clean up their data

Designing a business oriented, scalable solution for the future of NavVis IVION

What are the users requirements coming from the interviews?
The users should be allowed to interact with multiple scans to avoid repeating the same operation.
They want to be fast but precise
This operation should not be destructive, meaning if an error was done, there should be a fallback option.

Based on these criteria and the competitor’s analysis we focused our efforts on:

  • Orthographic camera implementation
  • Point cloud cluster selection among multiple scans
  • Crop view/bounding box
  • Saving and reverting any edits

The Orthographic camera in an industry standard that allows users more precise selection.
Point cloud cluster selection among multiple scans, will allow for faster and efficient workflow. By Hiding points and not deleting them we make the workflow non destructive.

Additionally Orthographic camera advantages can be introduced in other parts of the software IVION making it a very appealing development and value proposition.

We kept it simple, following our IVION product principle: “Scalability through simplicity” – Design for scalability while maintaining simplicity in administration.

List of requirements for point cloud cleanup after user interviews and discusssions with CSMs.
Priorised using MoSCoW, listed all the must-haves to give customers all the necessary tools to achieve their results keeping the “scalability through simplicity” product principle

Pairing up with stakeholders to iterate fast

The first rough concept and wireframe validation was conducted within the UX design team, to get a first check on the flow and coherence with current and upcoming features.

After that a more refined idea was introduced to the development team.
We used the initial development time to spike orthographic projection, switching between perspective and orthographic, and the different selection options.
Once these spikes were assigned, to further refine the concept, the approach that worked best for me and my team was to pair up with the “topic expert”, e.g. orthographic camera, point cloud rendering, etc, and design together.

This allowed to get buy-in and help discussing doubts in larger meetings, and since NavVis IVION is the gateway for companies needing reality capturing software, we aligned on few guiding principles:

  • In doubt analyse the competition, we want to maximise adoption by offering familiar tools.
  • Don’t invest time in redesigning other features than the ones discussed to maximise resources
  • Users can interact only on what is visible, if scans are hidden they should not be affected by any action
cloud compare and IVION point cloud cleanup concept

Comparison between cloud compare limit box and selection flow, and an high level concept for point cloud cleanup in ivion to present shown to tech leads.

User requirements concepts

User requirements concepts discussion with tech leads to address feasibility and scalability

First high fidelity design for point cloud cleanup

First finalised design for point cloud cleanup including user requirements, MVP for usability testing

Iterative approach to solution validation. Alpha and Beta testing.

The approach that we chose to validate the feature was iterative.
We internally defined two programs Alpha and Beta, were alpha is tested internally with industry experts and Customer success managers, and Beta is a feature flag limited testing run with customers.

Alpha is mandatory, but based on the results Beta could be skipped in favour of an earlier release.
We enlisted our point cloud cleanup experts (with competitors tools), to conduct 10 semi-structured, moderated usability testing interviews of around 60 to 90 minutes each to go over different key aspects.

We aimed at a CSAT score of around 85-90% to consider the alpha release ready.

And despite some additional insights and feedback to make the feature even better we still reached a score of 94%. This gave our C-suite enough confidence to release without a Beta phase.

We addressed the quick wins, prepared design fixes and the feature was deemed ready to release after a single bi-weekly sprint.

Feedback result from alpha usability testing with prioritised issues based on scoping and timeline.

Final high fidelity design PCC

New improved UI and UX based on the feedback received from the Alpha, additional features might be considered for future major releases.

Setting up a net to catch all the post release feedback

With the release of the new feature approaching, knowing the anticipation from our customers and the high expectations, I aligned with the customer facing teams and leadership to setup a feedback repository.

All the bugs and technical issues would keep on being collected and addressed by our support team in jira, while the UX and product feedback is first collected in confluence with thorough details, similar issues will be collected in the same spot labeled with the customer who faced this issue and the CSM who raised it.

Additionally we use mixpanel dashboards to track churn.

With the combination of qualitative and quantitative information we can more easily reach out to specific customers, if needed, and get more information on what is the problem they are trying to solve and how we can more efficiently address it with the right solution in the shortest amount of time.

Retrospective: Pair up to get buy-ins from stakeholders.

If there was a lesson that while working on this project, is this: pairing up with key stakeholders to address design concerns and issues, will go a long way in ensuring buy-ins and smooth development.

After learning that different members of the development team had a different understading, I invested individual time, using different visual examples from competitors, preparing prototypes but mostly understanding their communication style and getting on the same level, asking them to explain it back to me with their own words.

This is also a great practice to foster a good collaboration and feedback culture, as well as a “team building” activity.