Can I read EReFormer: Event-Based Depth Estimation on EtoBox?

EReFormer: Event-Based Depth Estimation by 張傑 is a document available to read on EtoBox.

What is EReFormer: Event-Based Depth Estimation about?

This paper presents EReFormer, a novel event-based monocular depth estimator that employs recurrent transformers to effectively model global spatial information and long-range temporal dependencies from event streams. The proposed architecture includes a transformer-based encoder-decoder, a Gate Recurrent Vision Transformer (GRViT) for enhanced temporal modeling, and a Cross Attention-guided Skip Connection (CASC) for improved spatial modeling. Experimental results demonstrate that EReFormer significantly o

Author

張傑

Language

EN