Questions about the difference between the real image size and annotated image size

[stdKonjac]

Sorry for the bother, but I found the real image size is inconsistent with that of annotation_FSC147_384.json. For example, the size of download image 1050.jpg is 469 x 384 (W x H), but the size in annotation_FSC147_384.json is 1300 x 1065. I wonder why these two shapes are different and how can I use the bbox in annotation_FSC147_384.json? @Viresh-R @v-hoainm

[Ranjan]

Hey,
we resized the images for the experiment. For 1050.jpg, 1300 * 1065 is the original size of the image, while 469*384 is the resized image size. The exemplars in the annotation file correspond to the resized image, so you can directly use them.
Note that we have shared the resized images.

[stdKonjac]

Thank you for your nice explanation!

[jaideep11061982]

@stdKonjac
were you able to map the bbx to image
I dont get right boxes on to the image
,how y1/y2 here are same ,143/143.

for b in [[143, 266, 143, 317]]:
    #@break 
    print(b)
    y,x,y1,x1=b
    #print(b)
    cv2.rectangle(img ,(x,y),(x1,y1),(255,0,0),5)

plt.imshow(img)

[stdKonjac]

@stdKonjac were you able to map the bbx to image I dont get right boxes on to the image ,how y1/y2 here are same ,143/143.

for b in [[143, 266, 143, 317]]:
    #@break 
    print(b)
    y,x,y1,x1=b
    #print(b)
    cv2.rectangle(img ,(x,y),(x1,y1),(255,0,0),5)

plt.imshow(img)

Perhaps the bbox is [left_top_x, left_top_y, h, w] instead of [x1, y1, x2, y2]?