2026.08
Our paper “Pair-Relative Partial Equivariance for Rotation-Robust Keypoint Descriptor Matching” has been accepted to the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2027.
The paper was accepted in Round 1 of the WACV 2027 review cycle.
A single-author work by Prof. Jongmin Lee. The paper introduces a pair-relative partial equivariance formulation for learning rotation-robust local keypoint descriptors, improving feature matching under large in-plane rotations.