nvinfer1::plugin::RPROIParams Struct Reference

RPROIParams is used to create the RPROIPlugin instance. More...

#include <NvInferPluginUtils.h>

Public Attributes
int32_t	poolingH
	Height of the output in pixels after ROI pooling on feature map. More...
int32_t	poolingW
	Width of the output in pixels after ROI pooling on feature map. More...
int32_t	featureStride
int32_t	preNmsTop
	Number of proposals to keep before applying NMS. More...
int32_t	nmsMaxOut
	Number of remaining proposals after applying NMS. More...
int32_t	anchorsRatioCount
	Number of anchor box ratios. More...
int32_t	anchorsScaleCount
	Number of anchor box scales. More...
float	iouThreshold
	IoU (Intersection over Union) threshold used for the NMS step. More...
float	minBoxSize
	Minimum allowed bounding box size before scaling, used for anchor box calculation. More...
float	spatialScale
	Spatial scale between the input image and the last feature map. More...

Detailed Description

RPROIParams is used to create the RPROIPlugin instance.

Member Data Documentation

anchorsRatioCount

int32_t nvinfer1::plugin::RPROIParams::anchorsRatioCount

Number of anchor box ratios.

anchorsScaleCount

int32_t nvinfer1::plugin::RPROIParams::anchorsScaleCount

Number of anchor box scales.

featureStride

int32_t nvinfer1::plugin::RPROIParams::featureStride

Feature stride; ratio of input image size to feature map size. Assuming that max pooling layers in the neural network use square filters.

iouThreshold

float nvinfer1::plugin::RPROIParams::iouThreshold

IoU (Intersection over Union) threshold used for the NMS step.

minBoxSize

float nvinfer1::plugin::RPROIParams::minBoxSize

Minimum allowed bounding box size before scaling, used for anchor box calculation.

nmsMaxOut

int32_t nvinfer1::plugin::RPROIParams::nmsMaxOut

Number of remaining proposals after applying NMS.

poolingH

int32_t nvinfer1::plugin::RPROIParams::poolingH

Height of the output in pixels after ROI pooling on feature map.

poolingW

int32_t nvinfer1::plugin::RPROIParams::poolingW

Width of the output in pixels after ROI pooling on feature map.

preNmsTop

int32_t nvinfer1::plugin::RPROIParams::preNmsTop

Number of proposals to keep before applying NMS.

spatialScale

float nvinfer1::plugin::RPROIParams::spatialScale

Spatial scale between the input image and the last feature map.

---

The documentation for this struct was generated from the following file:

• NvInferPluginUtils.h