Search papers, labs, and topics across Lattice.
The Univer- sity of Hong Kong
3
0
7
Foundation models are revolutionizing monocular depth estimation, achieving unprecedented accuracy and robustness in real-world applications.
By synthesizing diverse training data and leveraging pixel-wise calibration, FoundationGeo achieves unprecedented robustness in monocular metric geometry, even under varying camera intrinsics.
Models ace existing claim verification benchmarks by only checking the most obvious constraint, and fall apart when forced to consider multiple constraints together.
