Camera pose and spatial arrangement are the most important dimensions for both collecting diverse robot data and retrieving useful subsets, and aligned retrieval from DROID beats full-dataset co-training on real tasks.
Note that we do not ablate on material properties of these objects
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
What Matters in Learning from Large-Scale Datasets for Robot Manipulation
Camera pose and spatial arrangement are the most important dimensions for both collecting diverse robot data and retrieving useful subsets, and aligned retrieval from DROID beats full-dataset co-training on real tasks.