Abstract
3D visual illusion is a perceptual phenomenon where a two-dimensional planeis manipulated to simulate three-dimensional spatial relationships, making aflat artwork or object look three-dimensional in the human visual system. Inthis paper, we reveal that the machine visual system is also seriously fooledby 3D visual illusions, including monocular and binocular depth estimation. Inorder to explore and analyze the impact of 3D visual illusion on depthestimation, we collect a large dataset containing almost 3k scenes and 200kimages to train and evaluate SOTA monocular and binocular depth estimationmethods. We also propose a robust depth estimation framework that uses commonsense from a vision-language model to adaptively select reliable depth frombinocular disparity and monocular depth. Experiments show that SOTA monocular,binocular, and multi-view depth estimation approaches are all fooled by various3D visual illusions, while our method achieves SOTA performance.