ncnn_retinaface.h
2.2 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
//
// Created by DefTruth on 2021/11/20.
//
#ifndef LITE_AI_TOOLKIT_NCNN_CV_NCNN_RETINAFACE_H
#define LITE_AI_TOOLKIT_NCNN_CV_NCNN_RETINAFACE_H
#include "lite/ncnn/core/ncnn_core.h"
namespace ncnncv
{
class LITE_EXPORTS NCNNRetinaFace : public BasicNCNNHandler
{
public:
explicit NCNNRetinaFace(const std::string &_param_path,
const std::string &_bin_path,
unsigned int _num_threads = 1,
int _input_height = 640,
int _input_width = 640);
~NCNNRetinaFace() override = default;
private:
// nested classes
struct RetinaAnchor
{
float cx;
float cy;
float s_kx;
float s_ky;
};
private:
const int input_height; // 640/320
const int input_width; // 640/320
const float mean_vals[3] = {104.f, 117.f, 123.f}; // bgr order
const float norm_vals[3] = {1.f, 1.f, 1.f};
const float variance[2] = {0.1f, 0.2f};
std::vector<int> steps = {8, 16, 32};
std::vector<std::vector<int>> min_sizes = {
{16, 32},
{64, 128},
{256, 512}
};
enum NMS
{
HARD = 0, BLEND = 1, OFFSET = 2
};
static constexpr const unsigned int max_nms = 30000;
private:
void transform(const cv::Mat &mat, ncnn::Mat &in) override;
void generate_anchors(const int target_height,
const int target_width,
std::vector<RetinaAnchor> &anchors);
void generate_bboxes(std::vector<types::Boxf> &bbox_collection,
ncnn::Extractor &extractor,
float score_threshold, float img_height,
float img_width); // rescale & exclude
void nms(std::vector<types::Boxf> &input, std::vector<types::Boxf> &output,
float iou_threshold, unsigned int topk, unsigned int nms_type);
public:
void detect(const cv::Mat &mat, std::vector<types::Boxf> &detected_boxes,
float score_threshold = 0.7f, float iou_threshold = 0.3f,
unsigned int topk = 300, unsigned int nms_type = 0);
};
}
#endif //LITE_AI_TOOLKIT_NCNN_CV_NCNN_RETINAFACE_H