Is there any chance to replace SO3 and SE3 motor pattern expression in SLAM (eg vins-mono or ORB-SLAM3) in the future?

[naiixn]

Hello, I have read your article in detail. As a junior researcher, I am very interested in the innovation in your paper. Here I would like to ask whether SE2 (3) may replace the current popular one in the future, the SO3 or SE3 motion expression formulas in SLAM, and is the computational complexity of this motion expression formula large? Can it be done in real time if it is applied to SLAM? I would be greatly appreciated if I could get a reply from the author! If possible, I would like to contact you by email, my email: 1187772963@qq.com