|
ReUseX
0.0.5
3D Point Cloud Processing for Building Reuse
|
Data carrier for LibTorch YOLO inference pipeline. More...
#include <Data.hpp>


Public Attributes | |
| cv::Mat | image |
| Input: raw image (BGR, HWC). | |
| int | target_size = 640 |
| Letterbox target dimension. | |
| float | letterbox_scale = 1.0f |
| Scale applied during letterbox. | |
| cv::Size | original_size |
| Pre-letterbox dimensions. | |
| cv::Mat | label_image |
| Output: label image (CV_16U, class+1). | |
| float | confidence_threshold = 0.25f |
| Detection confidence threshold. | |
| float | iou_threshold = 0.45f |
| NMS IoU threshold. | |
| int | max_detections = 300 |
| Maximum detections per image. | |
Additional Inherited Members | |
| Public Member Functions inherited from reusex::vision::IData | |
| virtual | ~IData ()=default |
Data carrier for LibTorch YOLO inference pipeline.
Holds the input image, preprocessing parameters, and output label image for a single sample processed through the LibTorch backend.
| float reusex::vision::libtorch::LibTorchData::confidence_threshold = 0.25f |
| cv::Mat reusex::vision::libtorch::LibTorchData::image |
| float reusex::vision::libtorch::LibTorchData::iou_threshold = 0.45f |
| cv::Mat reusex::vision::libtorch::LibTorchData::label_image |
| float reusex::vision::libtorch::LibTorchData::letterbox_scale = 1.0f |
| int reusex::vision::libtorch::LibTorchData::max_detections = 300 |
| cv::Size reusex::vision::libtorch::LibTorchData::original_size |
| int reusex::vision::libtorch::LibTorchData::target_size = 640 |