mmr/tkDNN
Archived
1
0
Fork 0
Code Issues 46 Pull Requests 12 Releases 4 Activity

compute detections

Browse Source
This commit is contained in:
Francesco Gatti
2019-02-04 20:34:15 +00:00
parent 88097a3774
commit c8dea4668d
3 changed files with 140 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
include/Layer.h
+22 -2
View File
@@ -339,13 +339,33 @@ struct sortable_bbox {
class Yolo : public Layer {
public:
Yolo(Network *net, int classes, int num);
struct box {
float x, y, w, h;
};
typedef struct detection{
Yolo::box bbox;
int classes;
float *prob;
float *mask;
float objectness;
int sort_class;
};
Yolo(Network *net, int classes, int num, const char* fname_weights);
virtual ~Yolo();
virtual layerType_t getLayerType() { return LAYER_YOLO; };
int classes, num;
dnnType *mask_h, *mask_d; //anchors
dnnType *bias_h, *bias_d; //anchors
virtual dnnType* infer(dataDim_t &dim, dnnType* srcData);
int computeDetections(int w, int h, float thresh);
const int MAX_DETECTIONS = 256;
Yolo::detection *dets;
int detected;
};
/**
src/Yolo.cpp
+105 -1
View File
@@ -11,12 +11,28 @@
namespace tk { namespace dnn {
Yolo::Yolo(Network *net, int classes, int num) :
Yolo::detection *make_network_boxes(int nboxes, int classes) {
int i;
Yolo::detection *dets = (Yolo::detection*) calloc(nboxes, sizeof(Yolo::detection));
for(i = 0; i < nboxes; ++i){
dets[i].prob = (float*) calloc(classes, sizeof(float));
}
return dets;
}
Yolo::Yolo(Network *net, int classes, int num, const char* fname_weights) :
Layer(net) {
this->classes = classes;
this->num = num;
// load anchors
int seek = 0;
readBinaryFile(fname_weights, num, &mask_h, &mask_d);
seek += num;
readBinaryFile(fname_weights, 3*num, &bias_h, &bias_d);
// same
output_dim.n = input_dim.n;
output_dim.c = input_dim.c;
@@ -25,6 +41,9 @@ Yolo::Yolo(Network *net, int classes, int num) :
output_dim.l = input_dim.l;
checkCuda( cudaMalloc(&dstData, output_dim.tot()*sizeof(dnnType)) );
dets = make_network_boxes(MAX_DETECTIONS, classes);
detected = 0;
}
Yolo::~Yolo() {
@@ -39,6 +58,43 @@ int entry_index(int batch, int location, int entry,
entry*input_dim.w*input_dim.h + loc;
}
Yolo::box get_yolo_box(float *x, float *biases, int n, int index, int i, int j, int lw, int lh, int w, int h, int stride) {
Yolo::box b;
b.x = (i + x[index + 0*stride]) / lw;
b.y = (j + x[index + 1*stride]) / lh;
b.w = exp(x[index + 2*stride]) * biases[2*n] / w;
b.h = exp(x[index + 3*stride]) * biases[2*n+1] / h;
return b;
}
void correct_yolo_boxes(Yolo::detection *dets, int n, int w, int h, int netw, int neth, int relative)
{
int i;
int new_w=0;
int new_h=0;
if (((float)netw/w) < ((float)neth/h)) {
new_w = netw;
new_h = (h * netw)/w;
} else {
new_h = neth;
new_w = (w * neth)/h;
}
for (i = 0; i < n; ++i){
Yolo::box b = dets[i].bbox;
b.x = (b.x - (netw - new_w)/2./netw) / ((float)new_w/netw);
b.y = (b.y - (neth - new_h)/2./neth) / ((float)new_h/neth);
b.w *= (float)netw/new_w;
b.h *= (float)neth/new_h;
if(!relative){
b.x *= w;
b.w *= w;
b.y *= h;
b.h *= h;
}
dets[i].bbox = b;
}
}
dnnType* Yolo::infer(dataDim_t &dim, dnnType* srcData) {
@@ -58,4 +114,52 @@ dnnType* Yolo::infer(dataDim_t &dim, dnnType* srcData) {
return dstData;
}
int Yolo::computeDetections(int w, int h, float thresh) {
dnnType *predictions = new dnnType[output_dim.tot()];
checkCuda( cudaMemcpy(predictions, dstData, output_dim.tot()*sizeof(dnnType), cudaMemcpyDeviceToHost));
int relative = 1;
int lw = output_dim.w;
int lh = output_dim.h;
int netw = net->input_dim.w;
int neth = net->input_dim.h;
if (output_dim.n == 2) {
FatalError("BATCH of 2 not supported");
//avg_flipped_yolo(l);
}
int i,j,n;
int count = 0;
for (i = 0; i < lw*lh; ++i){
int row = i / lw;
int col = i % lw;
for(n = 0; n < num; ++n){
int obj_index = entry_index(0, n*lw*lh + i, 4, classes, input_dim, output_dim);
float objectness = predictions[obj_index];
if(objectness <= thresh) continue;
int box_index = entry_index(0, n*lw*lh + i, 0, classes, input_dim, output_dim);
dets[count].bbox = get_yolo_box(predictions, bias_h, mask_h[n], box_index, col, row, lw, lh, netw, neth, lw*lh);
dets[count].objectness = objectness;
dets[count].classes = classes;
for(j = 0; j < classes; ++j){
int class_index = entry_index(0, n*lw*lh + i, 4 + 1 + j, classes, input_dim, output_dim);
float prob = objectness*predictions[class_index];
dets[count].prob[j] = (prob > thresh) ? prob : 0;
}
++count;
if(count >= MAX_DETECTIONS)
FatalError("reach max boxes");
}
}
correct_yolo_boxes(dets, count, w, h, netw, neth, relative);
std::cout<<"DETECTED: "<<count<<"\n";
detected = count;
return count;
}
}}
tests/yolo3_berkeley/yolo3_berkeley.cpp
+13 -3
View File
@@ -61,6 +61,7 @@ const char *c78_bin = "../tests/yolo3_berkeley/layers/c78.bin";
const char *c79_bin = "../tests/yolo3_berkeley/layers/c79.bin";
const char *c80_bin = "../tests/yolo3_berkeley/layers/c80.bin";
const char *c81_bin = "../tests/yolo3_berkeley/layers/c81.bin";
const char *c82_bin = "../tests/yolo3_berkeley/layers/g82.bin";
const char *c84_bin = "../tests/yolo3_berkeley/layers/c84.bin";
const char *c87_bin = "../tests/yolo3_berkeley/layers/c87.bin";
const char *c88_bin = "../tests/yolo3_berkeley/layers/c88.bin";
@@ -69,6 +70,7 @@ const char *c90_bin = "../tests/yolo3_berkeley/layers/c90.bin";
const char *c91_bin = "../tests/yolo3_berkeley/layers/c91.bin";
const char *c92_bin = "../tests/yolo3_berkeley/layers/c92.bin";
const char *c93_bin = "../tests/yolo3_berkeley/layers/c93.bin";
const char *c94_bin = "../tests/yolo3_berkeley/layers/g94.bin";
const char *c96_bin = "../tests/yolo3_berkeley/layers/c96.bin";
const char *c99_bin = "../tests/yolo3_berkeley/layers/c99.bin";
const char *c100_bin = "../tests/yolo3_berkeley/layers/c100.bin";
@@ -77,6 +79,7 @@ const char *c102_bin = "../tests/yolo3_berkeley/layers/c102.bin";
const char *c103_bin = "../tests/yolo3_berkeley/layers/c103.bin";
const char *c104_bin = "../tests/yolo3_berkeley/layers/c104.bin";
const char *c105_bin = "../tests/yolo3_berkeley/layers/c105.bin";
const char *c106_bin = "../tests/yolo3_berkeley/layers/g106.bin";
const char *output_bins[3] = {
"../tests/yolo3_berkeley/debug/layer82_out.bin",
"../tests/yolo3_berkeley/debug/layer94_out.bin",
@@ -238,7 +241,7 @@ int main() {
tk::dnn::Conv2d c80 (&net,1024, 3, 3, 1, 1, 1, 1, c80_bin, true);
tk::dnn::Activation a80 (&net, tk::dnn::ACTIVATION_LEAKY);
tk::dnn::Conv2d c81 (&net, 45, 1, 1, 1, 1, 0, 0, c81_bin, false);
tk::dnn::Yolo yolo0 (&net, 10, 3);
tk::dnn::Yolo yolo0 (&net, 10, 3, c84_bin);
tk::dnn::Layer *m83_layers[1] = { &a79 };
tk::dnn::Route m83 (&net, m83_layers, 1);
@@ -262,7 +265,7 @@ int main() {
tk::dnn::Conv2d c92 (&net, 512, 3, 3, 1, 1, 1, 1, c92_bin, true);
tk::dnn::Activation a92 (&net, tk::dnn::ACTIVATION_LEAKY);
tk::dnn::Conv2d c93 (&net, 45, 1, 1, 1, 1, 0, 0, c93_bin, false);
tk::dnn::Yolo yolo1 (&net, 10, 3);
tk::dnn::Yolo yolo1 (&net, 10, 3, c94_bin);
tk::dnn::Layer *m95_layers[1] = { &a91 };
tk::dnn::Route m95 (&net, m95_layers, 1);
@@ -286,7 +289,7 @@ int main() {
tk::dnn::Conv2d c104 (&net, 256, 3, 3, 1, 1, 1, 1, c104_bin, true);
tk::dnn::Activation a104 (&net, tk::dnn::ACTIVATION_LEAKY);
tk::dnn::Conv2d c105 (&net, 45, 1, 1, 1, 1, 0, 0, c105_bin, false);
tk::dnn::Yolo yolo2 (&net, 10, 3);
tk::dnn::Yolo yolo2 (&net, 10, 3, c106_bin);
// Load input
dnnType *data;
@@ -317,6 +320,13 @@ int main() {
cudnn_out[0] = yolo0.dstData;
cudnn_out[1] = yolo1.dstData;
cudnn_out[2] = yolo2.dstData;
printCenteredTitle(" compute detections ", '=', 30);
TIMER_START
yolo0.computeDetections(640, 480, 0.5);
yolo1.computeDetections(640, 480, 0.5);
yolo2.computeDetections(640, 480, 0.5);
TIMER_STOP
tk::dnn::dataDim_t dim2 = dim;
printCenteredTitle(" TENSORRT inference ", '=', 30); {