Search papers, labs, and topics across Lattice.
4
0
6
16
Over 1,500 submissions revealed stark differences in model performance across diverse domains, highlighting the challenges of generalizing egocentric video understanding.
Prior knowledge about signal representations can significantly boost the performance of automatic modulation classification in shifting communication environments.
Maritime computer vision is making strides in real-time feasibility, as evidenced by the MaCVi 2026 challenge results.
Forget complex cross-modal alignment: this method uses visual prompting with instance IDs and reinforcement learning to achieve a 20.9% m_IoU improvement on spatial-temporal video grounding.
