Search papers, labs, and topics across Lattice.
The Chinese University of Hong Kong
3
0
4
Task-aware sampling can dramatically enhance model efficiency, enabling superior performance with fewer parameters in challenging imaging tasks.
Neural networks can now solve complex geometric mapping problems orders of magnitude faster than traditional PDE solvers, and without needing any labeled training data.
Achieve shape-aware deep learning by baking in geometric priors with a differentiable module that normalizes, predicts, and regularizes shapes using Harmonic Beltrami Signatures.
