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

Add mobilenetv2ssdlite512 test.

Browse Source 
Works for CuDNN, not for tensorRT.
Modified channels in second convolution for classification headers from 126 to 486, when changing size from 300 to 512.
Added size 512 SSD specs and support to COCO dataset (81 classes, first BACKGROUND due to repo for training).
Refactoring class names.

Signed-off-by: Micaela Verucchi <micaelaverucchi@gmail.com>
This commit is contained in:
Micaela Verucchi
2020-03-17 10:35:02 +01:00
parent 9c25d15ff2
commit 1d388ca51a
5 changed files with 637 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/MobilenetDetection.cpp
+77 -27
View File
@@ -15,7 +15,7 @@ void MobilenetDetection::generate_ssd_priors(const SSDSpec *specs, const int n_s
n_priors = 0;
for (int i = 0; i < n_specs; i++)
{
n_priors += specs[i].feature_size * specs[i].feature_size * 6;
n_priors += specs[i].featureSize * specs[i].featureSize * 6;
}
// std::cout<<"n priors: "<<n_priors<<std::endl;
@@ -28,18 +28,18 @@ void MobilenetDetection::generate_ssd_priors(const SSDSpec *specs, const int n_s
int min, max;
for (int i = 0; i < n_specs; i++)
{
scale = (float)image_size / (float)specs[i].shrinkage;
min = specs[i].box_height > specs[i].box_width ? specs[i].box_width : specs[i].box_height;
max = specs[i].box_height < specs[i].box_width ? specs[i].box_width : specs[i].box_height;
for (int j = 0; j < specs[i].feature_size; j++)
scale = (float)imageSize / (float)specs[i].shrinkage;
min = specs[i].boxHeight > specs[i].boxWidth ? specs[i].boxWidth : specs[i].boxHeight;
max = specs[i].boxHeight < specs[i].boxWidth ? specs[i].boxWidth : specs[i].boxHeight;
for (int j = 0; j < specs[i].featureSize; j++)
{
for (int k = 0; k < specs[i].feature_size; k++)
for (int k = 0; k < specs[i].featureSize; k++)
{
//small sized square box
size = min;
x_center = (k + 0.5f) / scale;
y_center = (j + 0.5f) / scale;
h = w = (float)size / (float)image_size;
h = w = (float)size / (float)imageSize;
priors[i_prio * N_COORDS + 0] = x_center;
priors[i_prio * N_COORDS + 1] = y_center;
@@ -49,7 +49,7 @@ void MobilenetDetection::generate_ssd_priors(const SSDSpec *specs, const int n_s
//big sized square box
size = sqrt(max * min);
h = w = (float)size / (float)image_size;
h = w = (float)size / (float)imageSize;
priors[i_prio * N_COORDS + 0] = x_center;
priors[i_prio * N_COORDS + 1] = y_center;
@@ -59,7 +59,7 @@ void MobilenetDetection::generate_ssd_priors(const SSDSpec *specs, const int n_s
//change h/w ratio of the small sized box
size = min;
h = w = size / (float)image_size;
h = w = size / (float)imageSize;
ratio = sqrt(specs[i].ratio1);
priors[i_prio * N_COORDS + 0] = x_center;
priors[i_prio * N_COORDS + 1] = y_center;
@@ -103,15 +103,15 @@ void MobilenetDetection::generate_ssd_priors(const SSDSpec *specs, const int n_s
}
}
void MobilenetDetection::convert_locatios_to_boxes_and_center(float *priors, const int n_priors, float *locations, const float center_variance, const float size_variance)
void MobilenetDetection::convert_locatios_to_boxes_and_center(float *priors, const int n_priors, float *locations, const float centerVariance, const float sizeVariance)
{
float cur_x, cur_y;
for (int i = 0; i < n_priors; i++)
{
locations[i * N_COORDS + 0] = locations[i * N_COORDS + 0] * center_variance * priors[i * N_COORDS + 2] + priors[i * N_COORDS + 0];
locations[i * N_COORDS + 1] = locations[i * N_COORDS + 1] * center_variance * priors[i * N_COORDS + 3] + priors[i * N_COORDS + 1];
locations[i * N_COORDS + 2] = exp(locations[i * N_COORDS + 2] * size_variance) * priors[i * N_COORDS + 2];
locations[i * N_COORDS + 3] = exp(locations[i * N_COORDS + 3] * size_variance) * priors[i * N_COORDS + 3];
locations[i * N_COORDS + 0] = locations[i * N_COORDS + 0] * centerVariance * priors[i * N_COORDS + 2] + priors[i * N_COORDS + 0];
locations[i * N_COORDS + 1] = locations[i * N_COORDS + 1] * centerVariance * priors[i * N_COORDS + 3] + priors[i * N_COORDS + 1];
locations[i * N_COORDS + 2] = exp(locations[i * N_COORDS + 2] * sizeVariance) * priors[i * N_COORDS + 2];
locations[i * N_COORDS + 3] = exp(locations[i * N_COORDS + 3] * sizeVariance) * priors[i * N_COORDS + 3];
cur_x = locations[i * N_COORDS + 0];
cur_y = locations[i * N_COORDS + 1];
@@ -221,16 +221,38 @@ float MobilenetDetection::get_color2(int c, int x, int max)
return r;
}
void MobilenetDetection::init(std::string tensor_path)
void MobilenetDetection::init(std::string tensor_path, int input_size, int n_classes)
{
this->imageSize = input_size;
this->classes = n_classes;
const int n_SSDSpec = 6;
SSDSpec specs[6];
specs[0].setAll(19, 16, 60, 105, 2, 3);
specs[1].setAll(10, 32, 105, 150, 2, 3);
specs[2].setAll(5, 64, 150, 195, 2, 3);
specs[3].setAll(3, 100, 195, 240, 2, 3);
specs[4].setAll(2, 150, 240, 285, 2, 3);
specs[5].setAll(1, 300, 285, 330, 2, 3);
if(input_size == 300)
{
specs[0].setAll(19, 16, 60, 105, 2, 3);
specs[1].setAll(10, 32, 105, 150, 2, 3);
specs[2].setAll(5, 64, 150, 195, 2, 3);
specs[3].setAll(3, 100, 195, 240, 2, 3);
specs[4].setAll(2, 150, 240, 285, 2, 3);
specs[5].setAll(1, 300, 285, 330, 2, 3);
}
else if(input_size == 512)
{
specs[0].setAll(32, 16, 60, 105, 2, 3);
specs[1].setAll(16, 32, 105, 150, 2, 3);
specs[2].setAll(8, 64, 150, 195, 2, 3);
specs[3].setAll(4, 100, 195, 240, 2, 3);
specs[4].setAll(2, 150, 240, 285, 2, 3);
specs[5].setAll(1, 300, 285, 330, 2, 3);
}
else
{
FatalError("Input size for mobilenet not supported");
}
generate_ssd_priors(specs, n_SSDSpec);
@@ -242,7 +264,7 @@ void MobilenetDetection::init(std::string tensor_path)
locations_h = (float *)malloc(N_COORDS * n_priors * sizeof(float));
confidences_h = (float *)malloc(n_priors * classes * sizeof(float));
dim = tk::dnn::dataDim_t(1, 3, input_w, input_h, 1);
dim = tk::dnn::dataDim_t(1, 3, imageSize, imageSize, 1);
for (int c = 0; c < classes; c++)
{
@@ -253,11 +275,39 @@ void MobilenetDetection::init(std::string tensor_path)
colors[c] = cv::Scalar(int(255.0 * b), int(255.0 * g), int(255.0 * r));
}
const char *voc_class_name_[] = {
if(classes == 21)
{
const char *classes_names_[] = {
"BACKGROUND", "aeroplane", "bicycle", "bird", "boat", "bottle", "bus",
"car", "cat", "chair", "cow", "diningtable", "dog", "horse", "motorbike",
"person", "pottedplant", "sheep", "sofa", "train", "tvmonitor"};
voc_class_name = std::vector<std::string>(voc_class_name_, std::end(voc_class_name_));
classesNames = std::vector<std::string>(classes_names_, std::end(classes_names_));
}
else if (classes == 81)
{
const char *classes_names_[] = {
"BACKGROUND", "person" , "bicycle" , "car" , "motorbike" , "aeroplane" , "bus" ,
"train" , "truck" , "boat" , "traffic light" , "fire hydrant" , "stop sign" ,
"parking meter" , "bench" , "bird" , "cat" , "dog" , "horse" , "sheep" , "cow" ,
"elephant" , "bear" , "zebra" , "giraffe" , "backpack" , "umbrella" , "handbag" ,
"tie" , "suitcase" , "frisbee" , "skis" , "snowboard" , "sports ball" , "kite" ,
"baseball bat" , "baseball glove" , "skateboard" , "surfboard" , "tennis racket" ,
"bottle" , "wine glass" , "cup" , "fork" , "knife" , "spoon" , "bowl" , "banana" ,
"apple" , "sandwich" , "orange" , "broccoli" , "carrot" , "hot dog" , "pizza" ,
"donut" , "cake" , "chair" , "sofa" , "pottedplant" , "bed" , "diningtable" ,
"toilet" , "tvmonitor" , "laptop" , "mouse" , "remote" , "keyboard" ,
"cell phone" , "microwave" , "oven" , "toaster" , "sink" , "refrigerator" ,
"book" , "clock" , "vase" , "scissors" , "teddy bear" , "hair drier" , "toothbrush"};
classesNames = std::vector<std::string>(classes_names_, std::end(classes_names_));
}
else
{
FatalError("Number of classes not supported for mobilenet");
}
}
cv::Mat MobilenetDetection::draw()
@@ -266,7 +316,7 @@ cv::Mat MobilenetDetection::draw()
for (size_t i = 0; i < detected.size(); i++)
{
b = detected[i];
std::string det_class = voc_class_name[b.cl];
std::string det_class = classesNames[b.cl];
cv::rectangle(origImg, cv::Point(b.x, b.y), cv::Point(b.w, b.h), colors[b.cl], 2);
// draw label
cv::Size textSize = getTextSize(det_class, cv::FONT_HERSHEY_SIMPLEX, fontScale, thickness, &baseline);
@@ -319,8 +369,8 @@ void MobilenetDetection::update(cv::Mat &img)
checkCuda(cudaMemcpy(locations_h, loc, N_COORDS * n_priors * sizeof(float), cudaMemcpyDeviceToHost));
//postprocess
convert_locatios_to_boxes_and_center(priors, n_priors, locations_h, center_variance, size_variance);
detected = postprocess(locations_h, confidences_h, n_priors, conf_thresh, classes, iou_threshold, sz.width, sz.height);
convert_locatios_to_boxes_and_center(priors, n_priors, locations_h, centerVariance, sizeVariance);
detected = postprocess(locations_h, confidences_h, n_priors, confThresh, classes, iouThreshold, sz.width, sz.height);
TIMER_STOP
stats.push_back(t_ns);