Measuring uncertainty in human visual segmentation

Segmenting visual stimuli into distinct groups of features and visual objects is central to visual function. Classical psychophysical methods have helped uncover many rules of human perceptual segmentation, and recent progress in machine learning has produced successful algorithms. Yet, the computational logic of human segmentation remains unclear, partially because we lack well-controlled paradigms to measure perceptual segmentation maps and compare models quantitatively. Here we propose a new, integrated approach: given an image, we measure multiple pixel-based same-different judgments and perform model-based reconstruction of the underlying segmentation map. The reconstruction is robust to several experimental manipulations and captures the variability of individual participants. We demonstrate the validity of the approach on human segmentation of natural images and composite textures. We show that image uncertainty affects measured human variability, and it influences how participants weigh different visual features. Because any putative segmentation algorithm can be inserted to perform the reconstruction, our paradigm affords quantitative tests of theories of perception as well as new benchmarks for segmentation algorithms. Visual segmentation is the process of decomposing the visual field into meaningful parts. Segmentation is the focus of a vast literature in visual perception and neuroscience, because it is a core function of the visual system that involves bottom/up and top/down integration across the whole visual cortex. Similarly, segmentation is an essential task of computer vision systems, because it is required for countless practical applications. However, the lack of rigorous empirical measures of segmentation-related uncertainty represents a major roadblock for both fields, because subjective uncertainty is a central feature of visual perception, and also because existing databases do not allow to calibrate segmentation algorithms that do compute uncertainty. The work presented in this manuscript proposes to overcome these limitations. Specifically, our contributions are threefold: (i) We introduce the first experimental method to measure perceptual segmentation on arbitrary images. (ii) We capture individual-level variability and relate it to perceptual uncertainty, which is necessary to understand human perception. (iii) We offer computational tools to fit any segmentation algorithm to the data, which will enable new benchmarks for computer vision algorithms, and testing computational theories of perceptual segmentation.