GstInference/Helper Elements/Detection Crop

GstInference
Introduction
Getting started
Getting the code Building the plugin Quick Starting Guide
Supported architectures
InceptionV1 InceptionV2 InceptionV3 InceptionV4 TinyYoloV2 TinyYoloV3 YoloV2 ResNetV1 MobileNetV2 MobileNetV2 + SSD AlexNet
Supported backends
TensorFlow Tensorflow-Lite Coral from Google Caffe TensorRT ONNXRT ONNXRT ACL ONNXRT OpenVINO
Metadata and Signals
GstClassificationMeta GstInferenceMeta Signals
Overlay Elements
Inference Overlay
Utils Elements
InferenceFilter InferenceCrop InferenceDebug InferenceBin
Legacy pipelines
PC NANO TX2 Xavier IMX8
Example pipelines
Example applications
Classification Detection Hand Pose Recognition
Benchmarks
Model Zoo
Project Status
Releases Roadmap
Contact Us

This element consumes the detection metadata introduced in the buffer by our family of detection elements(TinyYoloV2). This element uses GstDetectionMeta.

Setection crop example

Properties

The detectioncrop element expose the following properties to control the aspect ratio and the detection index. These properties are documented on the following table:

Property	Value	Description
crop-index	UInt [0,4294967295]	Index of the detected object to crop in the prediction array. This will be ignored if crop-class is set to a non-negative value.
crop-class	Int [-1,2147483647]	Object class to crop look for cropping. If set to -1, crop-index will be used.
aspect-ratio	fraction [ 0/1, 2147483647/1]	Aspect ratio to crop the detections, width and height separated by '/'. If set to 0/1 it maintains the aspect ratio of each bounding box.

Example

For example pipelines using detectioncrop element please check the example pipelines section.