A controlled empirical study with a unified architecture finds SE(3)-equivariant position-orientation convolutions outperform less constrained models on geometry-aligned tasks, and pose-based symmetry breaking gives consistent gains.
This is because the action of g ∈ SE(3) permutes these values on the fiber or spatially, but the activation acts on each scalar value independently
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
-
Probing Equivariance and Symmetry Breaking in Convolutional Networks
A controlled empirical study with a unified architecture finds SE(3)-equivariant position-orientation convolutions outperform less constrained models on geometry-aligned tasks, and pose-based symmetry breaking gives consistent gains.