Crowdsourcing 3D Semantic Maps for Vehicle Cognition Cognition for - PowerPoint PPT Presentation

Crowdsourcing 3D Semantic Maps for Vehicle Cognition

Cognition for Cars Decisions Eyes Cognition Civil Maps Cloud Bounding boxes Civil Maps In Car

Vehicle Cognition

Vehicle Cognition through 3D Maps

Vehicle Cognition through 3D Maps

Vehicle Cognition through 3D Maps

Vehicle Cognition through 3D Maps

Vehicle Cognition through 3D Maps

Vehicle Cognition through 3D Maps

Vehicle Cognition through 3D Maps

Vehicle Cognition through 3D Maps

Vehicle Cognition through 3D Maps

3D Semantic Maps

3D Semantic Maps

3D Semantic Maps to Cognition Maps Projected into a Maps Projected onto a Camera View Maps on Earth Field of View (requires precise localizaiton)

6D Localization

Sensor data -> 3D Maps & Localization 3D Semantic Map Localization Map Sensors Point Cloud

Problem: GPS is not reliable or accurate ● Actual vehicle GPS traces collected in downtown San Francisco

Localization Methods: ICP VS Signatures SLOW: ICP FAST: Signatures ● Point to point comparisons ● Signature to Signature comparisons ● 2 million points per second ● 1000 Signature per second ● Lots of CPU / GPU usage ● ARM Cortex is enough

Synthetics : AI Training Process Training 90% Procedural Synthetic 3D modeling Sensor Data 3D Semantic AI Models Maps Real World Real Sensor Driving Data Validation 10%

Crowdsourced 3D Semantic Maps

Crowdsourced Maps: Client/Server Semantic Aggregation Averaging, Hyp. Testing, Derivation (Completeness) Outlier rejection Job Dispatching Local Map ... Map Auditing ... Cache

Crowdsourced Maps: Supply & Demand Map Contributions Map Usage Map Aggregation Civil Maps

Free Webinar Series by Civil Maps (4 Sessions) Getting Started With Sensor Data Civil Maps Github: videos & code ○ https://civilmaps.github.io/cm-hal/ Session Three: (Sensor Fusion Part 1): Thursday, May 25th 1:30 PM PDT (4:30 PM EST) Session Four: (Sensor Fusion Part 2): Thursday, June 29th 1:30 PM PDT (4:30 PM EST)

Towards Mixed Reality Programmable Roads

Stay in Touch: Engineering Blog on Medium: https://medium.com/@CivilMaps

Contact Fabien Chraim, VP Research and Development fabien@civilmaps.com Scott Harvey, Sr. Machine Vision Engineer scott@civilmaps.com Cognition for Cars