Header: AVLDL.h
Namespace: avl
Module: DeepLearning

Executes a Locate Points model on a single input image.


void avl::DL_LocatePoints
	const avl::Image& inImage,
	atl::Optional<const avl::Region&> inRoi,
	const avl::LocatePointsModelId& inModelId,
	const float inMinDetectionScore,
	const atl::Optional<float>& inMinDistanceRatio,
	const bool inOverlap,
	atl::Array<avl::Location>& outLocations,
	atl::Array<int>& outClassIds,
	atl::Array<atl::String>& outClassNames,
	atl::Array<float>& outScores


Name Type Range Default Description
inImage const Image& Input image
inRoi Optional<const Region&> NIL Limits the area where points may be located
inModelId const LocatePointsModelId& Identifier of a Locate Points model
inMinDetectionScore const float 0.0 - 1.0 0.5f Sets a minimum required score for a point to be returned
inMinDistanceRatio const Optional<float>& 0.01 - 1.0 NIL Sets a minimum distance between the returned points defined as a portion of the Feature Size. If not set, a value determined during the training is used
inOverlap const bool True Cuts the image into more overlapping tiles, which improves results quality at the expense of extended execution time
outLocations Array<Location>& Returns location of the found points
outClassIds Array<int>& Returns ids of the found point classes
outClassNames Array<String>& Returns names of the found point classes
outScores Array<float>& Returns scores of the found points


For input inImage only pixel formats are supported: 1⨯uint8, 3⨯uint8.

Read more about pixel formats in Image documentation.


  • It is recommended that the deep learning model is deployed with DL_LocatePoints_Deploy first and connected through the inModelId input.
  • If one decides not to use DL_LocatePoints_Deploy, then the model will be loaded in the first iteration. It will take up to several seconds.
  • Use inOverlap=False to increase execution speed at a cost of lower precision of results.


This filter should not be executed along with running Deep Learning Service as it may result in degraded performance or even out-of-memory errors. However, if stopping Deep Learning Service is not an option (e.g. program uses other Deep Learning filters related to Instances Segmentation) it is advised that this filter should be executed at least once before performing operations utilizing Deep Learning Service (e.g. executing filters related to Instances Segmentation). Such "warm up" execution does not have to use real image or roi as long as provided image and roi have similar size.


List of possible exceptions:

Error type Description
DomainError Not supported inImage pixel format in DL_LocatePoints.

See Also

  • Models for Deep Learning may be created using Adaptive Vision Deep Learning Editor or using Training Api.