Attribution-guided object crops from geolocation models retain more predictive information than random crops of equal area on a three-country benchmark.
SAM 2: Segment Anything in Images and Videos,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2representative citing papers
BRIXEL applies self-distillation to let vision foundation models like DINOv3 produce higher-quality dense features at fixed input resolution.
citing papers explorer
-
Object-Level Explanations for Image Geolocation Models: a GeoGuessr use-case
Attribution-guided object crops from geolocation models retain more predictive information than random crops of equal area on a three-country benchmark.
-
Another BRIXEL in the Wall: Towards Cheaper Dense Features
BRIXEL applies self-distillation to let vision foundation models like DINOv3 produce higher-quality dense features at fixed input resolution.