How do I obtain the distance from the bounding boxes in YOLO v2/v3 in MATLAB?

Hello,

I currently trained my dataset on both YOLO v2 and v3; however, I am not sure on how to obtain the distance using the corners of the bounding boxes?

Is there a specific function or loop that I need to use?

I appreciate all the feedback.

[farah bouzidi]

did you find an answer to your issue?

Answer 1

[Vineet Joshi]

Hi

The standard bounding box format in YOLOv2/v3 is a four dimentional vector described as follows.

[x y width height]

Using this you can find the coordinate of the four corners of the bounding box.

Once you have all four coordinates you can use the pdist function to find the distances.

The following resources will help.

Pairwise distance between pairs of observations.

bboxes - Detect objects using YOLO v3 object detector.

Hope this helps

Thanks

Answer 2

[yanqi liu]

sir, may be use the dice to compute Sørensen-Dice similarity coefficient for image segmentation

clc; clear all; close all;
RGB = imread('yellowlily.jpg');
region1 = [350 700 425 120]; % [x y w h] format
BW1 = false(size(RGB,1),size(RGB,2));
BW1(region1(2):region1(2)+region1(4),region1(1):region1(1)+region1(3)) = true;

region2 = [800 1124 120 230];
BW2 = false(size(RGB,1),size(RGB,2));
BW2(region2(2):region2(2)+region2(4),region2(1):region2(1)+region2(3)) = true;

region3 = [20 1320 480 200; 1010 290 180 240]; 
BW3 = false(size(RGB,1),size(RGB,2));    
BW3(region3(1,2):region3(1,2)+region3(1,4),region3(1,1):region3(1,1)+region3(1,3)) = true;
BW3(region3(2,2):region3(2,2)+region3(2,4),region3(2,1):region3(2,1)+region3(2,3)) = true;

imshow(RGB)
hold on
visboundaries(BW1,'Color','r');
visboundaries(BW2,'Color','g');
visboundaries(BW3,'Color','b');
title('Seed Regions')

L = imseggeodesic(RGB,BW1,BW2,BW3,'AdaptiveChannelWeighting',true);

L_groundTruth = double(imread('yellowlily-segmented.png'));

similarity = dice(L, L_groundTruth)

similarity = 3×1

0.9396 0.7247 0.9139