大山的专栏

利用目标检测模型标注数据以构建xml文件代码如下：weights = 'pt/last.pt'det_model = ObjectDet(weights, input_size=(800, 800), conf_thres=0.5)data_dir = 'datasets/20220302/'names = ["two_wheel", "three_wheel", "car", "person"]folder_name = "20220302"database = "Unknown"fo

import numpy as npimport onnxfrom onnx import defs, checker, helper, numpy_helper, mappingfrom onnx import ModelProto, GraphProto, NodeProto, AttributeProto, TensorProto, OperatorProto, OperatorSetIdProtofrom onnx.helper import make_tensor, make_tenso

常用目标检测算法中如何从特征层中获取bbox和scorefloat *cls_score;float *box_pred;for (int j = 0; j < map_size; j++) { for (int k = 0; k < num_anchor; k++) { std::vector<float> sample_scores; for (int n = 0; n < num_classes; n++)

MMDetection config下模型配置文件解读faster_rcnn_r50_fpn_1x.py配置文件｀｀｀#model settingsmodel = dict(type=‘FasterRCNN’, # model类型pretrained=‘modelzoo://resnet50’, # 预训练模型：imagenet-resnet50backbone=dict(type=‘ResNet’,