High-quality and Scalable Multi-Sensor Fusion Data Annotation for Autonomous Mobility
Highly-accurate labeling of data collected from multiple sensors of an autonomous vehicle is crucial to improve the performance of computer vision models.
Cuboid annotation is simple. However, one cannot surround a 3D box on a 2D screen and annotate accurately. Our team leverages multiple viewpoints across projections for cuboid annotation. Once the cuboid has found a cluster of points, the annotator can swap between different views to annotate better.
Semantic segmentation of 3D point clouds is challenging, and the iMerit team has in-house tools to support the classification of spatial point clouds. With our tooling ecosystem of client tools, in-house and 3rd party tools, iMerit teams can remarkably annotate videos and images in separate frames for higher accuracy and efficiency.
Leading Autonomous Mobility Company partners with iMerit for LiDAR Annotation to build a 3D Perception System
We partnered with this company to support them with data annotation across 2D images and 3D point clouds. 3D perception systems are highly dependent on data quality for improved performance, and the company was looking at target identification in LiDAR frames with lane marking, road boundaries, traffic lights, and others.
With our human-in-the-loop workflows, data labeling on 3D LiDAR frames for poles, pedestrians, signs, cars, and barriers, was achieved seamlessly and accurately.
"iMerit's incredibly knowledgeable data labelers are part of our team and are the right people for our enrichment efforts."
Director of Data Products, Research Institute of Autonomous Driving
Autonomous vehicle companies are vigorously working towards building a robust system for efficiently tackling different situations that arise on the road.
One of the crucial steps to achieve this is anomaly detection and categorization. Generalizing what is observed in the training data will be counter-productive to building robust models.
Check out this webinar from iMerit’s ML Data Ops Summit, discussing anomaly detection and test-case generation.
