在Faster RCNN中常见的NMS是在rpn的最后一步,合并重复的IOU用的,由于RPN的预测只有前景和背景之分,所以目前我看到的代码中所用的NMS都是不是多类别的NMS,但是在YOLO这种网络中应该需要用到多类别的NMS。
function picks = nms_multiclass(boxes, overlap)
%%boxes为一个m*n的矩阵,其中m为boundingbox的个数,n的前4列为每个boundingbox的坐标,格式为
%%(x1,y1,x2,y2);第5:n列为每一类的置信度,一共n-5+1个置信度
% top = nms(boxes, overlap)
% Non-maximum suppression. (FAST VERSION)
% Greedily select high-scoring detections and skip detections
% that are significantly covered by a previously selected
% detection.
%
% NOTE: This is adapted from Pedro Felzenszwalb's version (nms.m),
% but an inner loop has been eliminated to significantly speed it
% up in the case of a large number of boxes
% Copyright (C) 2011-12 by Tomasz Malisiewicz
% All rights reserved.
%
% This file is part of the Exemplar-SVM library and is made
% available under the terms of the MIT license (see COPYING file).
% Project homepage: https://github.com/quantombone/exemplarsvm
if isempty(boxes)
picks = {};
return;
end
if size(boxes, 1) < 10000
picks = nms_multiclass_mex(double(boxes), double(overlap));
return;
end
x1 = boxes(:,1);
y1 = boxes(:,2);
x2 = boxes(:,3);
y2 = boxes(:,4);
area = (x2-x1+1) .* (y2-y1+1);
picks = cell(size(boxes, 2)-4, 1);
%在不同类别内分别做NMS
for iS = 5:size(boxes, 2)
s = boxes(:,iS);
[~, I] = sort(s);
pick = s*0;
counter = 1;
while ~isempty(I)
last = length(I);
i = I(last);
pick(counter) = i;
counter = counter + 1;
xx1 = max(x1(i), x1(I(1:last-1)));
yy1 = max(y1(i), y1(I(1:last-1)));
xx2 = min(x2(i), x2(I(1:last-1)));
yy2 = min(y2(i), y2(I(1:last-1)));
w = max(0.0, xx2-xx1+1);
h = max(0.0, yy2-yy1+1);
inter = w.*h;
o = inter ./ (area(i) + area(I(1:last-1)) - inter);
I = I(o<=overlap);
end
pick = pick(1:(counter-1));
picks{iS-4} = pick;
end