Search papers, labs, and topics across Lattice.
4
0
6
0
Unpredictable modality combinations can drastically hinder VI-ReID performance, but the Modality Adaptive Matching Transformer (MAMT) offers a robust solution by dynamically fusing features based on modality relationships.
DDTNet reveals that focusing on degradation transfer can dramatically improve image restoration performance across diverse weather conditions.
Current video models struggle to infer unseen spatial states and causal relationships, falling far short of human-level spatial reasoning.
Softmax's overconfidence in noisy-label person Re-ID can be tamed by adaptive similarity calibration and evidence propagation, leading to more robust feature learning.
