ML & Segmentation

The original ML message types were removed in 0.1.10. For ML inference pipelines, use:

Model I/O: Tensor and TensorPool for zero-copy tensor transport
Detection results: Detection / Detection3D with bounding boxes
Custom outputs: GenericMessage for cross-language data
Segmentation: SegmentationMask (below)

SegmentationMask

Output type for semantic, instance, and panoptic segmentation models. Fixed-size header (64 bytes) — the pixel data follows in shared memory.

use horus::prelude::*;

// Semantic segmentation (e.g., from DeepLab — 21 classes)
let mask = SegmentationMask::semantic(640, 480, 21)
    .with_frame_id("camera_front")
    .with_timestamp(1234567890);

// Instance segmentation (e.g., from Mask R-CNN)
let mask = SegmentationMask::instance(640, 480);

// Panoptic segmentation (e.g., from Panoptic-FPN — 80 classes)
let mask = SegmentationMask::panoptic(640, 480, 80);

// Send via topic
let topic: Topic<SegmentationMask> = Topic::new("segmentation")?;
topic.send(mask);

Fields:

Field	Type	Description
`width`	`u32`	Mask width in pixels
`height`	`u32`	Mask height in pixels
`num_classes`	`u32`	Number of semantic classes
`mask_type`	`u32`	0=semantic, 1=instance, 2=panoptic
`timestamp_ns`	`u64`	Nanoseconds since epoch
`seq`	`u64`	Sequence number
`frame_id`	`[u8; 32]`	Camera frame identifier

Methods:

Method	Returns	Description
`semantic(w, h, num_classes)`	`SegmentationMask`	Create semantic mask header
`instance(w, h)`	`SegmentationMask`	Create instance mask header
`panoptic(w, h, num_classes)`	`SegmentationMask`	Create panoptic mask header
`with_frame_id(id)`	`Self`	Set frame ID
`with_timestamp(ts)`	`Self`	Set timestamp
`frame_id()`	`&str`	Get frame ID as string
`data_size()`	`usize`	Mask data size in bytes (w * h)

ML & Segmentation

SegmentationMask

See Also