Jul 2026
AVA-VLM: Adaptive Visual Attention-Vision Language Model for In-the-Wild Construction Site Monitoring
AVA-VLM is proposed, an Adaptive Visual Attention-Vision Language Model that follows a human-inspired coarse-to-fine strategy: it first reasons over a low-resolution global image and requests a high-resolution local crop only when detailed inspection is needed.
Young-Gwi Kim, Taeheon Kim, Youngseo Kim et al.
· arXiv.org · 0 citations