A unified model that predicts long-term coverage gains and obstacle maps, plus a new Doom-based benchmark, improves active 3D mapping efficiency in indoor scenes.
We study the impact of different spatial range information used to predict the next best path by training four different models on the AiMDoom Normal level training split
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
NextBestPath: Efficient 3D Mapping of Unseen Environments
A unified model that predicts long-term coverage gains and obstacle maps, plus a new Doom-based benchmark, improves active 3D mapping efficiency in indoor scenes.