Re: Bag of Binary Words VS RTABMap
Posted by leo152 on
URL: http://official-rtab-map-forum.206.s1.nabble.com/Bag-of-Binary-Words-VS-RTABMap-tp2335p2350.html
Hi!
Thanks for your response.
I found that in the Benchmark example, the official output of rtabmap-console of CityCentre, it cost 1300s to process 1237 images. Is that because the two options "-DbSqlite3/InMemory true -Mem/BinDataKept false"?
When processing the CityCentre dataset, timeLikelihoodCalculation cost 100ms per image and timePosteriorCalculation cost 50ms per image. As a result, it may cost 150ms per image to calculate loop closure detection?
In the original paper of BoBW, it claimes that Bag of words processing time is below 20ms when processing 26292 images as shown below. It may cost 30ms to deal with features, but only 20ms per image to process loop closure detection in a large dataset(26292 images).
URL: http://official-rtab-map-forum.206.s1.nabble.com/Bag-of-Binary-Words-VS-RTABMap-tp2335p2350.html
Hi!
Thanks for your response.
I found that in the Benchmark example, the official output of rtabmap-console of CityCentre, it cost 1300s to process 1237 images. Is that because the two options "-DbSqlite3/InMemory true -Mem/BinDataKept false"?
When processing the CityCentre dataset, timeLikelihoodCalculation cost 100ms per image and timePosteriorCalculation cost 50ms per image. As a result, it may cost 150ms per image to calculate loop closure detection?
In the original paper of BoBW, it claimes that Bag of words processing time is below 20ms when processing 26292 images as shown below. It may cost 30ms to deal with features, but only 20ms per image to process loop closure detection in a large dataset(26292 images).
| Free forum by Nabble | Edit this page |