Family 01
Depth & 3D
Monocular depth and stereo matching over the cameras and baseline you chose — dense 3D without a locked vendor stereo kit, plus automatic self-calibration to hold accuracy as the rig drifts.
Invotet SDK
Describe your rig. Run the pipeline.
A Python SDK for configuring the RVPU — describe your cameras and their geometry, pick the perception functions you need, and stream 3D positions, depth maps, and tracks into ROS 2. It ingests PyTorch, ONNX, and HuggingFace models when a pipeline needs a custom detector, with no CUDA in the loop. App config, not model code.
Frameworks
Bring a custom detector when a pipeline needs one
Framework
PyTorch
Bring a custom detector: trace or torch.export models fold into a pipeline with no rewrite.
Framework
ONNX
Standards-based interchange — any ONNX-exported model can back a pipeline stage.
Framework
HuggingFace
Vision checkpoints load through a one-line loader when a pipeline is customised.
A few lines, one pipeline
Describe the rig. Connect the RVPU. Stream.
Describe the cameras you chose and how they are mounted, pick the perception functions you need, and stream the results into ROS 2. Everything runs on-platform through a deterministic runtime — no model to train, no CUDA in the loop.
invotet-sdk · v0.1.0
from invotet import RVPU, Rig, Pipeline
# Describe the cameras you chose and how they are mounted
rig = Rig.stereo(camera="ov9281", baseline_mm=120, resolution=(1280, 720))
# Pick perception functions; the unit runs them on-platform
with RVPU.connect() as device:
device.load(Pipeline("dense-3d", "detect-track", rig=rig))
device.stream(ros2_topic="/perception")Pipelines supported
Vision pipelines, on the platform that has to see for itself.
Family 02
Detection & tracking
Object detection and segmentation, characterization of selected object classes, 3D mapping, and temporal tracking — structured perception for planners and manipulators.
Family 03
Cross-modality & scene
Thermal–RGB matching with follow-on inference on matched regions, scene parsing, and multi-sensor fusion into one on-device representation.
Become a design partner.
The SDK is available through our design-partner program while the toolchain matures. Tell us your platform, the pipelines you need, and the integration target — we will get you set up.