TI Deep Learning Product User Guide
sTIDL_DetectOutputParams_t Struct Reference

Detailed Description

This structure define the parameters of Detection Output Layer in TIDL.

Data Fields

int32_t processingType
 
int32_t priorBox
 
int32_t priorBoxSize
 
int32_t numClasses
 
int32_t backgroundLabelId
 
int32_t codeType
 
float32_tidl confThreshold
 
float32_tidl nmsThreshold
 
float32_tidl eta
 
int32_t topK
 
int32_t keepTopK
 
int32_t shareLocation
 
int32_t varianceEncoded
 
int32_t numKeypoints
 
int32_t keypointConfidence
 
int32_t numHeads
 
int32_t imWidth
 
int32_t imHeight
 
float32_tidl point3dRange [3][2]
 
float32_tidl voxelSize [2]
 
int32_t maxPointsPerVoxel
 
int32_t scoreConverter
 
int32_t metaArchType
 
int32_t dataLayout
 

Field Documentation

◆ processingType

int32_t sTIDL_DetectOutputParams_t::processingType

Processing type to be done refer eTIDL_processingType

◆ priorBox

int32_t sTIDL_DetectOutputParams_t::priorBox

Buffer containing the data required to form prior Bboxs

◆ priorBoxSize

int32_t sTIDL_DetectOutputParams_t::priorBoxSize

Size of the priorBox buffer required to form prior Bboxs

◆ numClasses

int32_t sTIDL_DetectOutputParams_t::numClasses

number of classes to be detected in the detection Output

◆ backgroundLabelId

int32_t sTIDL_DetectOutputParams_t::backgroundLabelId

To indicate whether or not to ignore background class

◆ codeType

int32_t sTIDL_DetectOutputParams_t::codeType

Indicates the coding type to be used for decoding Bboxs

◆ confThreshold

float32_tidl sTIDL_DetectOutputParams_t::confThreshold

Value to indicates threshold above which objects to be considered for detection

◆ nmsThreshold

float32_tidl sTIDL_DetectOutputParams_t::nmsThreshold

Threshold Value used for finding overlap between the bboxs in the NMS

◆ eta

float32_tidl sTIDL_DetectOutputParams_t::eta

Value used to update the adaptive Threshold in the NMS

◆ topK

int32_t sTIDL_DetectOutputParams_t::topK

Number of top k objects to keep for class after applying NMS

◆ keepTopK

int32_t sTIDL_DetectOutputParams_t::keepTopK

Number of top k objects to Keep in the final output

◆ shareLocation

int32_t sTIDL_DetectOutputParams_t::shareLocation

Indicate whether same size Boxes used for all classes or not, it is not supported in ti_dl

◆ varianceEncoded

int32_t sTIDL_DetectOutputParams_t::varianceEncoded

Flag to indicate the variance used in decoding bboxes is encoded along with locations are not

◆ numKeypoints

int32_t sTIDL_DetectOutputParams_t::numKeypoints

Number of keyPoints to be detected along with box

◆ keypointConfidence

int32_t sTIDL_DetectOutputParams_t::keypointConfidence

Flag to indicate whether there is an associated confidence for each key-point

◆ numHeads

int32_t sTIDL_DetectOutputParams_t::numHeads

Num heads used by the Post processing layer

◆ imWidth

int32_t sTIDL_DetectOutputParams_t::imWidth

Base Image Width used in 2D OD

◆ imHeight

int32_t sTIDL_DetectOutputParams_t::imHeight

Base Image Height usedin 2D OD

◆ point3dRange

float32_tidl sTIDL_DetectOutputParams_t::point3dRange[3][2]

3d data range(min-max) in x,y,z direction. Used in 3D OD

◆ voxelSize

float32_tidl sTIDL_DetectOutputParams_t::voxelSize[2]

voxel size in cm

◆ maxPointsPerVoxel

int32_t sTIDL_DetectOutputParams_t::maxPointsPerVoxel

Maximum number of 3D points per voxel

◆ scoreConverter

int32_t sTIDL_DetectOutputParams_t::scoreConverter

◆ metaArchType

int32_t sTIDL_DetectOutputParams_t::metaArchType

Meta Arch Type eTIDL_metaArchType

◆ dataLayout

int32_t sTIDL_DetectOutputParams_t::dataLayout