paolodonadei sfmh1 issues

paolodonadei / sfmh1

Automatically exported from code.google.com/p/sfmh1

0 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

robustness against meaningless Fs

#28 GoogleCodeExporter opened 8 years ago
0
crashing on valbonne sequence at the end

#27 GoogleCodeExporter opened 8 years ago
1
self calibration sucks

#26 GoogleCodeExporter opened 8 years ago
3
the program crashes when reading the oxford matches for the house and corridor sequence

#25 GoogleCodeExporter opened 8 years ago
0
corridor sequence not working

#24 GoogleCodeExporter opened 8 years ago
2
conditioning fundamental matrices before self calibration like how peter sturm does it, this might improve numerical stability

#23 GoogleCodeExporter opened 8 years ago
0
surf

#22 GoogleCodeExporter opened 8 years ago
1
OUTLIERS, OUTLIERS fml

#21 GoogleCodeExporter opened 8 years ago
6
additional matches

#20 GoogleCodeExporter opened 8 years ago
0
surviving dominant planes

#19 GoogleCodeExporter opened 8 years ago
0
texture maps

#18 GoogleCodeExporter opened 8 years ago
0
some times the index of a match is bigger than the number of available matches

#17 GoogleCodeExporter opened 8 years ago
1
crash when using house sequence

#16 GoogleCodeExporter opened 8 years ago
1
sift pca sometimes crashes

#15 GoogleCodeExporter opened 8 years ago
0
matches

#14 GoogleCodeExporter opened 8 years ago
1
distortion

#13 GoogleCodeExporter opened 8 years ago
0
bad matches remaining

#12 GoogleCodeExporter opened 8 years ago
0
he reconstruction is wrong for some initial frames

#11 GoogleCodeExporter opened 8 years ago
4
move the featuretrack class to the right file (matching) then change the correspondence class to an image relaton classs that includes geometry info in it

#10 GoogleCodeExporter opened 8 years ago
0
create trackmatrix class with overloaded [] operators and stop not abstracting out your classes

#9 GoogleCodeExporter opened 8 years ago
0
creation o fa routine that shwos matches across an arbitrary set of matches, this could be a video or a large pic

#8 GoogleCodeExporter opened 8 years ago
0
writing the trackmatrix to file so we can compare after and before pruning

#7 GoogleCodeExporter opened 8 years ago
0
augmented reality for unoredered image collections

#6 GoogleCodeExporter opened 8 years ago
0
use random kd trees for matching

#5 GoogleCodeExporter opened 8 years ago
0
use kd trees for matching, or bin the features and match across a large number of images, read the David Lowe paper on fast ANN matching via trees

#4 GoogleCodeExporter opened 8 years ago
0
use graphiz like the paper for photosynth to visualize the matches in spring form graphs

#3 GoogleCodeExporter opened 8 years ago
0
match across all images, so far we only have match per pair, but we want match tracks for the set

#2 GoogleCodeExporter opened 8 years ago
0
add in matches using a much faster but less accurate technique

#1 GoogleCodeExporter opened 8 years ago
0