summaryrefslogtreecommitdiffstats
path: root/SD-VBS/benchmarks/sift/src/matlab/siftmatch.m
diff options
context:
space:
mode:
Diffstat (limited to 'SD-VBS/benchmarks/sift/src/matlab/siftmatch.m')
-rw-r--r--SD-VBS/benchmarks/sift/src/matlab/siftmatch.m60
1 files changed, 60 insertions, 0 deletions
diff --git a/SD-VBS/benchmarks/sift/src/matlab/siftmatch.m b/SD-VBS/benchmarks/sift/src/matlab/siftmatch.m
new file mode 100644
index 0000000..d4398c1
--- /dev/null
+++ b/SD-VBS/benchmarks/sift/src/matlab/siftmatch.m
@@ -0,0 +1,60 @@
1% SIFTMATCH Match SIFT features
2% MATCHES=SIFTMATCH(DESCR1, DESCR2) matches the two sets of SIFT
3% descriptors DESCR1 and DESCR2.
4%
5% The function uses the same algorithm suggested by D. Lowe [1] to
6% reject matches that are too ambiguous.
7%
8% SIFTMATCH(DESCR1, DESCR2, THRESH) uses [1] with the specified
9% threshold THRESH. A descriptor D1 is matched to a descriptor D2
10% only if the distance d(D1,D2) multiplied by THRESH is not greather
11% than the distance of D1 to all other descriptors. The default
12% value of THRESH is 1.5.
13%
14% The storage class of the descriptors can be either DOUBLE, FLOAT,
15% INT8 or UINT8. Usually interger classes are faster.
16%
17% [1] D. G. Lowe,
18% `Distinctive image features from scale-invariant keypoints,'
19% IJCV, vol. 2, no. 60, pp. 91–110, 2004.
20%
21% See also SIFT(), SIFTDESCRIPTOR().
22
23% AUTORIGHTS
24% Copyright (c) 2006 The Regents of the University of California.
25% All Rights Reserved.
26%
27% Created by Andrea Vedaldi
28% UCLA Vision Lab - Department of Computer Science
29%
30% Permission to use, copy, modify, and distribute this software and its
31% documentation for educational, research and non-profit purposes,
32% without fee, and without a written agreement is hereby granted,
33% provided that the above copyright notice, this paragraph and the
34% following three paragraphs appear in all copies.
35%
36% This software program and documentation are copyrighted by The Regents
37% of the University of California. The software program and
38% documentation are supplied "as is", without any accompanying services
39% from The Regents. The Regents does not warrant that the operation of
40% the program will be uninterrupted or error-free. The end-user
41% understands that the program was developed for research purposes and
42% is advised not to rely exclusively on the program for any reason.
43%
44% This software embodies a method for which the following patent has
45% been issued: "Method and apparatus for identifying scale invariant
46% features in an image and use of same for locating an object in an
47% image," David G. Lowe, US Patent 6,711,293 (March 23,
48% 2004). Provisional application filed March 8, 1999. Asignee: The
49% University of British Columbia.
50%
51% IN NO EVENT SHALL THE UNIVERSITY OF CALIFORNIA BE LIABLE TO ANY PARTY
52% FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES,
53% INCLUDING LOST PROFITS, ARISING OUT OF THE USE OF THIS SOFTWARE AND
54% ITS DOCUMENTATION, EVEN IF THE UNIVERSITY OF CALIFORNIA HAS BEEN
55% ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. THE UNIVERSITY OF
56% CALIFORNIA SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, BUT NOT
57% LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
58% A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS ON AN "AS IS"
59% BASIS, AND THE UNIVERSITY OF CALIFORNIA HAS NO OBLIGATIONS TO PROVIDE
60% MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.