What is missing: The 3D output (point_cloud) is currently too large for efficient routing in complex dataflows, especially when multiple cameras are used or when the network is a bottleneck.
Where: Data processing and output logic.
Observable effect: Users can toggle between FULL, HALF, and ROI (Region of Interest) modes. In ROI mode, the point cloud size is reduced by ~90%, as the node only emits 3D data for the specific volumes where objects have been detected.
**What is missing:** The 3D output (`point_cloud`) is currently too large for efficient routing in complex dataflows, especially when multiple cameras are used or when the network is a bottleneck.
**Where:** Data processing and output logic.
**Observable effect:** Users can toggle between `FULL`, `HALF`, and `ROI` (Region of Interest) modes. In `ROI` mode, the point cloud size is reduced by ~90%, as the node only emits 3D data for the specific volumes where objects have been detected.
cristhian
added this to the v0.1.0 - Reusable Vision Engine milestone 2026-03-24 20:44:58 +00:00
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
What is missing: The 3D output (
point_cloud) is currently too large for efficient routing in complex dataflows, especially when multiple cameras are used or when the network is a bottleneck.Where: Data processing and output logic.
Observable effect: Users can toggle between
FULL,HALF, andROI(Region of Interest) modes. InROImode, the point cloud size is reduced by ~90%, as the node only emits 3D data for the specific volumes where objects have been detected.