My new pySLAM v2 is out!

I am excited to release pySLAM v2. The new version allows you to play with SLAM techniques, visual-odometry, keyframes, bundle-adjustment, feature-matching, and many modern local features (based on new Deep Learning approaches). It was a long work to make everything accessible from a single python environment but it was worth it!

Now, you can easily test how many modern local features perform and compare with respect to classical ones, within a pure feature matching application or a within a SLAM/Visual-odometry pipeline. At the present time, the following feature detectors are supported:

• FAST
• Good features to track
• ORB
• ORB2 (improvements of ORB-SLAM2 to ORB detector)
• SIFT
• SURF
• KAZE
• AKAZE
• BRISK
• AGAST
• MSER
• StarDector/CenSurE
• Harris-Laplace
• SuperPoint
• D2-Net
• DELF
• Contextdesc
• LFNet
• R2D2
• Key.Net

The following feature descriptors are supported:

• ORB
• SIFT
• ROOT SIFT
• SURF
• AKAZE
• BRISK
• FREAK
• SuperPoint
• Tfeat
• BOOST_DESC
• DAISY
• LATCH
• LUCID
• VGG
• Hardnet
• GeoDesc
• SOSNet
• L2Net
• Log-polar descriptor
• D2-Net
• DELF
• Contextdesc
• LFNet
• R2D2

You can find more information in the file feature_types.py.

A serious benchmarking is still a work in progress. 

Source code: 
https://github.com/luigifreda/pyslam

Check my related tweet with people’s comments, etc.

I am excited to release pySLAM v2. Now you can play with #SLAM techniques, #VisualOdometry, #Keyframes, #BundleAdjustment, #FeatureMatching, and many modern #LocalFeatures (based on DL). Everything is accessible from a single #python environment. https://t.co/XEPiuvnXJG pic.twitter.com/6K5yi5Z2G1

— Luigi Freda (@LuigiFreda) May 4, 2020