Point Cloud Processing Jobs
Computer vision roles requiring Point Cloud Processing expertise, across all industries and experience levels.
Open Positions
What is Point Cloud Processing?
Point cloud processing works directly on unordered sets of 3D points from LiDAR, depth cameras or photogrammetry — registering, filtering, segmenting and classifying them. The irregular, permutation-invariant structure of the data makes it genuinely different from image processing.
Where Point Cloud Processing is used
LiDAR perception in vehicles, robotic bin-picking and navigation, as-built verification in construction, and forestry and crop volume estimation from aerial scans.
Roles that ask for Point Cloud Processing
- 3D Vision Engineer
- LiDAR Perception Engineer
- Perception Engineer
- Robotics Engineer
- Point Cloud Engineer
Related skills & tools
Point Cloud Processing jobs — common questions
Which networks handle point clouds directly?
PointNet and PointNet++ introduced permutation-invariant processing of raw points. Voxel-based methods such as VoxelNet and SECOND, and pillar-based ones like PointPillars, trade some fidelity for speed and remain the practical choice in automotive perception.
What is ICP and why does it come up constantly?
Iterative Closest Point aligns two point clouds by repeatedly matching nearest neighbours and solving for the transform that minimises their distance. It underpins registration, scan matching and LiDAR odometry, and its failure modes are a standard interview topic.
Do I need LiDAR hardware experience?
It helps considerably. Understanding scan patterns, motion distortion, intensity returns and the differences between spinning, solid-state and flash sensors separates engineers who can debug real data from those who only work on curated benchmarks.