Rashed Karim Bipul

Spatio-Temporal Masked Autoencoders with Contrastive Cross-Attention for Robust Video Semantic Segmentation in Adverse Autonomous Driving Conditions

ST-MaskContrast incorporates a decoupled space-time token masking strategy coupled with a multi-scale cross-attention temporal decoder that enforces invariant representations between clean and corrupted sequential frames via contrastive patch-level loss, providing unprecedented perceptual resilience for safety-critical...