Darknet: Why the bboxes have a coordinate offsets with python interface?

[AaronYKing]

I test the model of yolov3 with python interface of darknet.py. But the results of bboxes have a coordinate offsets. The result picture link is here. The code is as follow:

from ctypes import *
import math
import random
import cv2

def sample(probs):
    s = sum(probs)
    probs = [a/s for a in probs]
    r = random.uniform(0, 1)
    for i in range(len(probs)):
        r = r - probs[i]
        if r <= 0:
            return i
    return len(probs)-1

def c_array(ctype, values):
    arr = (ctype*len(values))()
    arr[:] = values
    return arr

class BOX(Structure):
    _fields_ = [("x", c_float),
                ("y", c_float),
                ("w", c_float),
                ("h", c_float)]

class DETECTION(Structure):
    _fields_ = [("bbox", BOX),
                ("classes", c_int),
                ("prob", POINTER(c_float)),
                ("mask", POINTER(c_float)),
                ("objectness", c_float),
                ("sort_class", c_int)]


class IMAGE(Structure):
    _fields_ = [("w", c_int),
                ("h", c_int),
                ("c", c_int),
                ("data", POINTER(c_float))]

class METADATA(Structure):
    _fields_ = [("classes", c_int),
                ("names", POINTER(c_char_p))]


#lib = CDLL("/home/pjreddie/documents/darknet/libdarknet.so", RTLD_GLOBAL)
lib = CDLL("libdarknet.so", RTLD_GLOBAL)
lib.network_width.argtypes = [c_void_p]
lib.network_width.restype = c_int
lib.network_height.argtypes = [c_void_p]
lib.network_height.restype = c_int

predict = lib.network_predict
predict.argtypes = [c_void_p, POINTER(c_float)]
predict.restype = POINTER(c_float)

set_gpu = lib.cuda_set_device
set_gpu.argtypes = [c_int]

make_image = lib.make_image
make_image.argtypes = [c_int, c_int, c_int]
make_image.restype = IMAGE

get_network_boxes = lib.get_network_boxes
get_network_boxes.argtypes = [c_void_p, c_int, c_int, c_float, c_float, POINTER(c_int), c_int, POINTER(c_int)]
get_network_boxes.restype = POINTER(DETECTION)

make_network_boxes = lib.make_network_boxes
make_network_boxes.argtypes = [c_void_p]
make_network_boxes.restype = POINTER(DETECTION)

free_detections = lib.free_detections
free_detections.argtypes = [POINTER(DETECTION), c_int]

free_ptrs = lib.free_ptrs
free_ptrs.argtypes = [POINTER(c_void_p), c_int]

network_predict = lib.network_predict
network_predict.argtypes = [c_void_p, POINTER(c_float)]

reset_rnn = lib.reset_rnn
reset_rnn.argtypes = [c_void_p]

load_net = lib.load_network
load_net.argtypes = [c_char_p, c_char_p, c_int]
load_net.restype = c_void_p

do_nms_obj = lib.do_nms_obj
do_nms_obj.argtypes = [POINTER(DETECTION), c_int, c_int, c_float]

do_nms_sort = lib.do_nms_sort
do_nms_sort.argtypes = [POINTER(DETECTION), c_int, c_int, c_float]

free_image = lib.free_image
free_image.argtypes = [IMAGE]

letterbox_image = lib.letterbox_image
letterbox_image.argtypes = [IMAGE, c_int, c_int]
letterbox_image.restype = IMAGE

load_meta = lib.get_metadata
lib.get_metadata.argtypes = [c_char_p]
lib.get_metadata.restype = METADATA

load_image = lib.load_image_color
load_image.argtypes = [c_char_p, c_int, c_int]
load_image.restype = IMAGE

rgbgr_image = lib.rgbgr_image
rgbgr_image.argtypes = [IMAGE]

predict_image = lib.network_predict_image
predict_image.argtypes = [c_void_p, IMAGE]
predict_image.restype = POINTER(c_float)

def detect(net, meta, image, thresh=.5, hier_thresh=.5, nms=.45):
    im = load_image(image, 0, 0)
    num = c_int(0)
    pnum = pointer(num)
    predict_image(net, im)
    dets = get_network_boxes(net, im.w, im.h, thresh, hier_thresh, None, 0, pnum)
    num = pnum[0]
    if (nms): do_nms_obj(dets, num, meta.classes, nms);

    res = []
    for j in range(num):
        for i in range(meta.classes):
            if dets[j].prob[i] > 0:
                b = dets[j].bbox
                res.append((meta.names[i], dets[j].prob[i], (b.x, b.y, b.w, b.h)))
                print 'libdarknet.so:',b.x, b.y, b.w, b.h
    res = sorted(res, key=lambda x: -x[1])
    free_image(im)
    free_detections(dets, num)
    return res

if __name__ == "__main__":
    net = load_net("cfg/yolov3.cfg", "yolov3.weights", 0)
    meta = load_meta("cfg/coco.data")
    r = detect(net, meta, "data/dog.jpg")
    print r 

    image = cv2.imread('data/dog.jpg')
    print image.shape

    n = 0
    for i in range(len(r)):
        n+=1
        cv2.rectangle(image, (int(r[i][2][0]), int(r[i][2][1])), (int(r[i][2][0]+r[i][2][2]), int(r[i][2][1]+r[i][2][3])), (0,255,0), 2)
    print 'dets num:', n
    cv2.imwrite('result_dog.jpg', image)
    cv2.imshow('result_dog', image)
    cv2.waitKey(0) 

What's the problem about it?

Any help will be grateful!

[PeterQuinn925]

Looks like the upper left corner of the bbox is actually the center. Move
the box coordinates by -w/2 and -h/2.

On Tue, Aug 14, 2018 at 8:40 AM AaronYKing [e-mail] wrote:

I test the model of yolov3 with python interface of darknet.py. But the
results of bboxes have a coordinate offsets. The result picture link is
here https://github.com/AaronYKing/BUG/blob/master/result_dog.jpg. The
code is as follow:

from ctypes import *
import math
import random
import cv2

def sample(probs):
s = sum(probs)
probs = [a/s for a in probs]
r = random.uniform(0, 1)
for i in range(len(probs)):
r = r - probs[i]
if r <= 0:
return i
return len(probs)-1

def c_array(ctype, values):
arr = (ctype*len(values))()
arr[:] = values
return arr

class BOX(Structure):
_fields_ = [("x", c_float),
("y", c_float),
("w", c_float),
("h", c_float)]

class DETECTION(Structure):
_fields_ = [("bbox", BOX),
("classes", c_int),
("prob", POINTER(c_float)),
("mask", POINTER(c_float)),
("objectness", c_float),
("sort_class", c_int)]

class IMAGE(Structure):
_fields_ = [("w", c_int),
("h", c_int),
("c", c_int),
("data", POINTER(c_float))]

class METADATA(Structure):
_fields_ = [("classes", c_int),
("names", POINTER(c_char_p))]

lib = CDLL("/home/pjreddie/documents/darknet/libdarknet.so", RTLD_GLOBAL)

lib = CDLL("libdarknet.so", RTLD_GLOBAL)
lib.network_width.argtypes = [c_void_p]
lib.network_width.restype = c_int
lib.network_height.argtypes = [c_void_p]
lib.network_height.restype = c_int

predict = lib.network_predict
predict.argtypes = [c_void_p, POINTER(c_float)]
predict.restype = POINTER(c_float)

set_gpu = lib.cuda_set_device
set_gpu.argtypes = [c_int]

make_image = lib.make_image
make_image.argtypes = [c_int, c_int, c_int]
make_image.restype = IMAGE

get_network_boxes = lib.get_network_boxes
get_network_boxes.argtypes = [c_void_p, c_int, c_int, c_float, c_float, POINTER(c_int), c_int, POINTER(c_int)]
get_network_boxes.restype = POINTER(DETECTION)

make_network_boxes = lib.make_network_boxes
make_network_boxes.argtypes = [c_void_p]
make_network_boxes.restype = POINTER(DETECTION)

free_detections = lib.free_detections
free_detections.argtypes = [POINTER(DETECTION), c_int]

free_ptrs = lib.free_ptrs
free_ptrs.argtypes = [POINTER(c_void_p), c_int]

network_predict = lib.network_predict
network_predict.argtypes = [c_void_p, POINTER(c_float)]

reset_rnn = lib.reset_rnn
reset_rnn.argtypes = [c_void_p]

load_net = lib.load_network
load_net.argtypes = [c_char_p, c_char_p, c_int]
load_net.restype = c_void_p

do_nms_obj = lib.do_nms_obj
do_nms_obj.argtypes = [POINTER(DETECTION), c_int, c_int, c_float]

do_nms_sort = lib.do_nms_sort
do_nms_sort.argtypes = [POINTER(DETECTION), c_int, c_int, c_float]

free_image = lib.free_image
free_image.argtypes = [IMAGE]

letterbox_image = lib.letterbox_image
letterbox_image.argtypes = [IMAGE, c_int, c_int]
letterbox_image.restype = IMAGE

load_meta = lib.get_metadata
lib.get_metadata.argtypes = [c_char_p]
lib.get_metadata.restype = METADATA

load_image = lib.load_image_color
load_image.argtypes = [c_char_p, c_int, c_int]
load_image.restype = IMAGE

rgbgr_image = lib.rgbgr_image
rgbgr_image.argtypes = [IMAGE]

predict_image = lib.network_predict_image
predict_image.argtypes = [c_void_p, IMAGE]
predict_image.restype = POINTER(c_float)

def detect(net, meta, image, thresh=.5, hier_thresh=.5, nms=.45):
im = load_image(image, 0, 0)
num = c_int(0)
pnum = pointer(num)
predict_image(net, im)
dets = get_network_boxes(net, im.w, im.h, thresh, hier_thresh, None, 0, pnum)
num = pnum[0]
if (nms): do_nms_obj(dets, num, meta.classes, nms);

res = []
for j in range(num):
    for i in range(meta.classes):
        if dets[j].prob[i] > 0:
            b = dets[j].bbox
            res.append((meta.names[i], dets[j].prob[i], (b.x, b.y, b.w, b.h)))
            print 'libdarknet.so:',b.x, b.y, b.w, b.h
res = sorted(res, key=lambda x: -x[1])
free_image(im)
free_detections(dets, num)
return res

if __name__ == "__main__":
net = load_net("cfg/yolov3.cfg", "yolov3.weights", 0)
meta = load_meta("cfg/coco.data")
r = detect(net, meta, "data/dog.jpg")
print r

image = cv2.imread('data/dog.jpg')
print image.shape

n = 0
for i in range(len(r)):
    n+=1
    cv2.rectangle(image, (int(r[i][2][0]), int(r[i][2][1])), (int(r[i][2][0]+r[i][2][2]), int(r[i][2][1]+r[i][2][3])), (0,255,0), 2)
print 'dets num:', n
cv2.imwrite('result_dog.jpg', image)
cv2.imshow('result_dog', image)
cv2.waitKey(0)

What's the problem about it?

Any help will be grateful!

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
https://github.com/pjreddie/darknet/issues/1055, or mute the thread
https://github.com/notifications/unsubscribe-auth/ARocJafbS-2C_vmjLdikBeuyZvmAYA7zks5uQu9WgaJpZM4V8rvC
.

>

Peter Quinn
[phone] (cell)

[AaronYKing]

Yes. Thx @PeterQuinn925

[eesaeedkarimi]

change
res.append((meta.names[i], dets[j].prob[i], (b.x, b.y, b.w, b.h)))
to
res.append((meta.names[i], dets[j].prob[i], (b.x - b.w / 2, b.y - b.h /2, b.w, b.h)))