CraterBench-R is a new retrieval benchmark where self-supervised ViTs with a training-free instance-token aggregation method achieve high accuracy for identifying individual craters while reducing storage needs.
GroupViT: Semantic segmentation emerges from text supervision
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
CraterBench-R: Instance-Level Crater Retrieval for Planetary Scale
CraterBench-R is a new retrieval benchmark where self-supervised ViTs with a training-free instance-token aggregation method achieve high accuracy for identifying individual craters while reducing storage needs.