An end-to-end fully parametric method for image-goal navigation that leverages self-supervised and manifold learning to replace the topological graph with a geodesic regressor.
MIT License
8
stars
1
forks
source link
Cleaned unused parameters, renamed instances to match paper's names #1
Cleaner version of the code where models match the same convention as in the paper plus unused parameters were pruned from the configs and class constructors.
Cleaner version of the code where models match the same convention as in the paper plus unused parameters were pruned from the configs and class constructors.