CrossJEPA: Cross-Modal Joint-Embedding Predictive Architecture for Efficient 3D Representation Learning from 2D Images
Under review, 2025
This manuscript is about extending JEPA to cross-modal for better 3D point cloud understanding.
Under review, 2025
This manuscript is about extending JEPA to cross-modal for better 3D point cloud understanding.
Published in WACV 2026
This paper is about generalizing splatting-based reconstruction kernels beyond the exponential family to a broader class of functions.
Under review, 2025
This manuscript is about an interdisciplinary AI prototype fusing social procurement in construction and AI.