unMORE learns three object-centric representations (existence, center field, boundary distance) and uses a network-free reasoning algorithm to discover multiple objects in real-world images, outperforming prior unsupervised methods.
The Settings #1/#2 are fairly comparable with our Settings #1/#2, whereas its Setting #3 is from the original paper
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
-
unMORE: Unsupervised Multi-Object Segmentation via Center-Boundary Reasoning
unMORE learns three object-centric representations (existence, center field, boundary distance) and uses a network-free reasoning algorithm to discover multiple objects in real-world images, outperforming prior unsupervised methods.