We find that the highest resolution in existing multimodal benchmarks is only 2K. To address the current lack of high-resolution multimodal benchmarks, we construct HR-Bench. HR-Bench consists two sub-tasks: Fine-grained Single-instance Perception (FSP) and Fine-grained Cross-instance Perception (FCP).… See the full description on the dataset page:
https://huggingface.co/datasets/DreamMr/HR-Bench.
