DISK is a differentiable sparse kernel decomposition method that approximates spatially-variant complex convolutions using optimized sparse samples, initialization for non-convex shapes, and interpolation, achieving higher fidelity than simulated annealing and lower cost than low-rank methods.
Moving mobile graphics
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.GR 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
DISK: Differentiable Sparse Kernel Complex for Efficient Spatially-Variant Convolution
DISK is a differentiable sparse kernel decomposition method that approximates spatially-variant complex convolutions using optimized sparse samples, initialization for non-convex shapes, and interpolation, achieving higher fidelity than simulated annealing and lower cost than low-rank methods.