Skip to main content

impl_yolo_split_segdet_float

edgefirst_decoder::yolo

Function impl_yolo_split_segdet_float 

Source 
pub fn impl_yolo_split_segdet_float<B: BBoxTypeTrait, BOX: Float + AsPrimitive<f32> + Send + Sync, SCORE: Float + AsPrimitive<f32> + Send + Sync, MASK: Float + AsPrimitive<f32> + Send + Sync, PROTO: Float + AsPrimitive<f32> + Send + Sync>(
    boxes_tensor: ArrayView2<'_, BOX>,
    scores_tensor: ArrayView2<'_, SCORE>,
    mask_tensor: ArrayView2<'_, MASK>,
    protos: ArrayView3<'_, PROTO>,
    score_threshold: f32,
    iou_threshold: f32,
    nms: Option<Nms>,
    output_boxes: &mut Vec<DetectBox>,
    output_masks: &mut Vec<Segmentation>,
)
where
    f32: AsPrimitive<SCORE>,
Expand description

Internal implementation of YOLO split detection segmentation decoding for float tensors.

Expected shapes of inputs:

  • boxes_tensor: (4, num_boxes)
  • scores_tensor: (num_classes, num_boxes)
  • mask_tensor: (num_protos, num_boxes)
  • protos: (proto_height, proto_width, num_protos)

§Panics

Panics if shapes don’t match the expected dimensions.