EndoVis-18-VQLA

The 'EndoVis-18-VQLA' dataset/benchmark contains annotated surgical images and corresponding visual questions and answers, and it is used to evaluate Visual Question Localized-Answering (VQLA) capabilities in robotic surgery.

Papers using EndoVis-18-VQLA (3)

TemporalDoRA: Temporal PEFT for Robust Surgical Video Question Answering2026Surgical-MambaLLM: Mamba2-enhanced Multimodal Large Language Model for VQLA in Robotic Surgery2025Surgical-LVLM: Learning to Adapt Large Vision-Language Model for Grounded Visual Question Answering in Robotic Surgery2024 · 6 cites