You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
@gaoxiang12 Hi Xiang, I tried running LDSO on TUM mono dataset sequence 02 and 09, and found that no Loop closure was detected. The parameter "setting_enableLoopClosing" in Setting.cc remained as true and I've checked that LoopClosing::CorrectLoop did execute when running LDSO. Then I used function FeatureMatcher::DrawMatches to visualize the matched features just after line 161 in LoopClosing::CorrectLoop, and discovered that there are many wrong feature matchings like the following figure shows. So I guess that's probably why no loop closure was detected. Could you please have a check and see if you also get a similar observation
? Thanks!
The text was updated successfully, but these errors were encountered:
@gaoxiang12 Hi Xiang, I tried running LDSO on TUM mono dataset sequence 02 and 09, and found that no Loop closure was detected. The parameter "setting_enableLoopClosing" in Setting.cc remained as true and I've checked that LoopClosing::CorrectLoop did execute when running LDSO. Then I used function FeatureMatcher::DrawMatches to visualize the matched features just after line 161 in LoopClosing::CorrectLoop, and discovered that there are many wrong feature matchings like the following figure shows. So I guess that's probably why no loop closure was detected. Could you please have a check and see if you also get a similar observation
? Thanks!
The text was updated successfully, but these errors were encountered: