mirror/darknet
1
0
Fork 0
mirror of https://github.com/pjreddie/darknet.git synced 2023-08-10 21:13:14 +03:00
Code Issues Projects Releases Wiki Activity

sides of box instead of coords

Browse Source
This commit is contained in:
Joseph Redmon 2015-03-30 11:09:01 -07:00
parent 6553b3f0e3
commit ad59d7e68e
3 changed files with 64 additions and 53 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

87
src/data.c
View File

@ -16,6 +16,7 @@ struct load_args{
int w; int w;
int nh; int nh;
int nw; int nw;
int num_boxes;
int jitter; int jitter;
int classes; int classes;
int background; int background;
@ -106,56 +107,62 @@ void randomize_boxes(box *b, int n)
} }
} }
void fill_truth_detection(char *path, float *truth, int classes, int height, int width, int num_height, int num_width, int dy, int dx, int jitter, int flip, int background) void fill_truth_detection(char *path, float *truth, int classes, int num_boxes, int flip, int background, float dx, float dy, float sx, float sy)
{ {
int box_height = height/num_height;
int box_width = width/num_width;
char *labelpath = find_replace(path, "VOC2012/JPEGImages", "labels"); char *labelpath = find_replace(path, "VOC2012/JPEGImages", "labels");
labelpath = find_replace(labelpath, ".jpg", ".txt"); labelpath = find_replace(labelpath, ".jpg", ".txt");
int count = 0; int count = 0;
box *boxes = read_boxes(labelpath, &count); box *boxes = read_boxes(labelpath, &count);
randomize_boxes(boxes, count); randomize_boxes(boxes, count);
float x, y, h, w; float l,r,t,b;
int id; int id;
int i; int i;
if(background){ if(background){
for(i = 0; i < num_height*num_width*(4+classes+background); i += 4+classes+background){ for(i = 0; i < num_boxes*num_boxes*(4+classes+background); i += 4+classes+background){
truth[i] = 1; truth[i] = 1;
} }
} }
for(i = 0; i < count; ++i){ for(i = 0; i < count; ++i){
x = boxes[i].x; l = boxes[i].left;
y = boxes[i].y; r = boxes[i].right;
w = boxes[i].w; t = boxes[i].top;
h = boxes[i].h; b = boxes[i].bottom;
id = boxes[i].id; id = boxes[i].id;
if(flip) x = 1-x;
x *= width + jitter;
y *= height + jitter;
x -= dx;
y -= dy;
int i = x/box_width;
int j = y/box_height;
if(i < 0) i = 0; if(flip){
if(i >= num_width) i = num_width-1; float left = l;
if(j < 0) j = 0; float right = r;
if(j >= num_height) j = num_height-1; r = 1-left;
l = 1-right;
}
l = l*sx-dx;
r = r*sx-dx;
t = t*sy-dy;
b = b*sy-dy;
float dw = constrain(0,1, (x - i*box_width)/box_width ); float x = (l+r)/2.;
float dh = constrain(0,1, (y - j*box_height)/box_height ); float y = (t+b)/2.;
float th = constrain(0,1, h*(height+jitter)/height );
float tw = constrain(0,1, w*(width+jitter)/width );
int index = (i+j*num_width)*(4+classes+background); if (x < 0 || x >= 1 || y < 0 || y >= 1) continue;
int i = (int)(x*num_boxes);
int j = (int)(y*num_boxes);
l = constrain(0, 1, l);
r = constrain(0, 1, r);
t = constrain(0, 1, t);
b = constrain(0, 1, b);
int index = (i+j*num_boxes)*(4+classes+background);
if(truth[index+classes+background+2]) continue; if(truth[index+classes+background+2]) continue;
if(background) truth[index++] = 0; if(background) truth[index++] = 0;
truth[index+id] = 1; truth[index+id] = 1;
index += classes; index += classes;
truth[index++] = dh; truth[index++] = l;
truth[index++] = dw; truth[index++] = r;
truth[index++] = th; truth[index++] = t;
truth[index++] = tw; truth[index++] = b;
} }
free(boxes); free(boxes);
} }
@ -272,23 +279,30 @@ void free_data(data d)
} }
} }
data load_data_detection_jitter_random(int n, char **paths, int m, int classes, int h, int w, int nh, int nw, int jitter, int background) data load_data_detection_jitter_random(int n, char **paths, int m, int classes, int h, int w, int num_boxes, int jitter, int background)
{ {
char **random_paths = get_random_paths(paths, n, m); char **random_paths = get_random_paths(paths, n, m);
int i; int i;
data d; data d;
d.shallow = 0; d.shallow = 0;
d.X = load_image_paths(random_paths, n, h, w); d.X = load_image_paths(random_paths, n, h, w);
int k = nh*nw*(4+classes+background); int k = num_boxes*num_boxes*(4+classes+background);
d.y = make_matrix(n, k); d.y = make_matrix(n, k);
for(i = 0; i < n; ++i){ for(i = 0; i < n; ++i){
int dx = rand()%jitter; int px = rand()%jitter;
int dy = rand()%jitter; px = 0;
int py = rand()%jitter;
py = 0;
float sy = (float) h / (h-jitter);
float sx = (float) w / (w-jitter);
float dy = (float) py / (h-jitter);
float dx = (float) px / (w-jitter);
int flip = rand()%2; int flip = rand()%2;
fill_truth_detection(random_paths[i], d.y.vals[i], classes, h-jitter, w-jitter, nh, nw, dy, dx, jitter, flip, background); fill_truth_detection(random_paths[i], d.y.vals[i], classes, num_boxes, flip, background, dx, dy, sx, sy);
image a = float_to_image(h, w, 3, d.X.vals[i]); image a = float_to_image(h, w, 3, d.X.vals[i]);
if(flip) flip_image(a); if(flip) flip_image(a);
jitter_image(a,h-jitter,w-jitter,dy,dx); jitter_image(a, h-jitter, w-jitter, py, px);
} }
d.X.cols = (h-jitter)*(w-jitter)*3; d.X.cols = (h-jitter)*(w-jitter)*3;
free(random_paths); free(random_paths);
@ -299,7 +313,7 @@ void *load_detection_thread(void *ptr)
{ {
printf("Loading data: %d\n", rand()); printf("Loading data: %d\n", rand());
struct load_args a = *(struct load_args*)ptr; struct load_args a = *(struct load_args*)ptr;
*a.d = load_data_detection_jitter_random(a.n, a.paths, a.m, a.classes, a.h, a.w, a.nh, a.nw, a.jitter, a.background); *a.d = load_data_detection_jitter_random(a.n, a.paths, a.m, a.classes, a.h, a.w, a.num_boxes, a.jitter, a.background);
translate_data_rows(*a.d, -128); translate_data_rows(*a.d, -128);
scale_data_rows(*a.d, 1./128); scale_data_rows(*a.d, 1./128);
free(ptr); free(ptr);
@ -317,6 +331,7 @@ pthread_t load_data_detection_thread(int n, char **paths, int m, int classes, in
args->w = w; args->w = w;
args->nh = nh; args->nh = nh;
args->nw = nw; args->nw = nw;
args->num_boxes = nw;
args->classes = classes; args->classes = classes;
args->jitter = jitter; args->jitter = jitter;
args->background = background; args->background = background;

2
src/data.h
View File

@ -21,7 +21,7 @@ data load_data(char **paths, int n, int m, char **labels, int k, int h, int w);
pthread_t load_data_thread(char **paths, int n, int m, char **labels, int k, int h, int w, data *d); pthread_t load_data_thread(char **paths, int n, int m, char **labels, int k, int h, int w, data *d);
pthread_t load_data_detection_thread(int n, char **paths, int m, int classes, int h, int w, int nh, int nw, int jitter, int background, data *d); pthread_t load_data_detection_thread(int n, char **paths, int m, int classes, int h, int w, int nh, int nw, int jitter, int background, data *d);
data load_data_detection_jitter_random(int n, char **paths, int m, int classes, int h, int w, int nh, int nw, int jitter, int background); data load_data_detection_jitter_random(int n, char **paths, int m, int classes, int h, int w, int num_boxes, int jitter, int background);
data load_data_image_pathfile(char *filename, char **labels, int k, int h, int w); data load_data_image_pathfile(char *filename, char **labels, int k, int h, int w);
data load_cifar10_data(char *filename); data load_cifar10_data(char *filename);

28
src/detection.c
View File

@ -27,12 +27,11 @@ void draw_detection(image im, float *box, int side)
float blue = get_color(2,class,classes); float blue = get_color(2,class,classes);
j += classes; j += classes;
int d = im.w/side; int left = box[j] *im.w;
int y = r*d+box[j]*d; int right = box[j+1]*im.w;
int x = c*d+box[j+1]*d; int top = box[j+2]*im.h;
int h = box[j+2]*im.h; int bot = box[j+3]*im.h;
int w = box[j+3]*im.w; draw_box(im, left, top, right, bot, red, green, blue);
draw_box(im, x-w/2, y-h/2, x+w/2, y+h/2,red,green,blue);
} }
} }
} }
@ -138,17 +137,14 @@ void validate_detection(char *cfgfile, char *weightfile)
for(j = 0; j < pred.rows; ++j){ for(j = 0; j < pred.rows; ++j){
for(k = 0; k < pred.cols; k += classes+4+background+nuisance){ for(k = 0; k < pred.cols; k += classes+4+background+nuisance){
float scale = 1.; float scale = 1.;
if(nuisance) scale = 1.-pred.vals[j][k]; if (nuisance) scale = 1.-pred.vals[j][k];
for(class = 0; class < classes; ++class){ for (class = 0; class < classes; ++class){
int index = (k)/(classes+4+background+nuisance);
int r = index/7;
int c = index%7;
int ci = k+classes+background+nuisance; int ci = k+classes+background+nuisance;
float y = (r + pred.vals[j][ci + 0])/7.; float left = pred.vals[j][ci + 0];
float x = (c + pred.vals[j][ci + 1])/7.; float right = pred.vals[j][ci + 1];
float h = pred.vals[j][ci + 2]; float top = pred.vals[j][ci + 2];
float w = pred.vals[j][ci + 3]; float bot = pred.vals[j][ci + 3];
printf("%d %d %f %f %f %f %f\n", (i-1)*m/splits + j, class, scale*pred.vals[j][k+class+background+nuisance], y, x, h, w); printf("%d %d %f %f %f %f %f\n", (i-1)*m/splits + j, class, scale*pred.vals[j][k+class+background+nuisance], left, right, top, bot);
} }
} }
} }