3D Visual Illusion Depth Estimation

TL;DR

This paper investigates how 3D visual illusions deceive machine depth estimation systems, demonstrating that current state-of-the-art methods are fooled, and proposes a new framework leveraging vision language models for improved depth estimation.

Contribution

The paper reveals the vulnerability of machine depth estimation to 3D visual illusions and introduces a novel adaptive fusion framework using vision language models.

Findings

SOTA depth estimation methods are fooled by 3D visual illusions

A large dataset of nearly 3k scenes and 200k images was collected for evaluation

The proposed framework achieves state-of-the-art performance in depth estimation under illusions

Abstract

3D visual illusion is a perceptual phenomenon where a two-dimensional plane is manipulated to simulate three-dimensional spatial relationships, making a flat artwork or object look three-dimensional in the human visual system. In this paper, we reveal that the machine visual system is also seriously fooled by 3D visual illusions, including monocular and binocular depth estimation. In order to explore and analyze the impact of 3D visual illusion on depth estimation, we collect a large dataset containing almost 3k scenes and 200k images to train and evaluate SOTA monocular and binocular depth estimation methods. We also propose a 3D visual illusion depth estimation framework that uses common sense from the vision language model to adaptively fuse depth from binocular disparity and monocular depth. Experiments show that SOTA monocular, binocular, and multi-view depth estimation approaches are all fooled by various 3D visual illusions, while our method achieves SOTA performance.