Faces as Lighting Probes via Unsupervised Deep Highlight Extraction
This provides a more precise lighting estimation method for computer vision applications like augmented reality, though it is incremental as it builds on existing highlight-based approaches.
The paper tackles the problem of estimating detailed scene illumination from a single image using human faces, achieving state-of-the-art performance by extracting highlights via an unsupervised deep neural network and tracing them to produce a non-parametric environment map.
We present a method for estimating detailed scene illumination using human faces in a single image. In contrast to previous works that estimate lighting in terms of low-order basis functions or distant point lights, our technique estimates illumination at a higher precision in the form of a non-parametric environment map. Based on the observation that faces can exhibit strong highlight reflections from a broad range of lighting directions, we propose a deep neural network for extracting highlights from faces, and then trace these reflections back to the scene to acquire the environment map. Since real training data for highlight extraction is very limited, we introduce an unsupervised scheme for finetuning the network on real images, based on the consistent diffuse chromaticity of a given face seen in multiple real images. In tracing the estimated highlights to the environment, we reduce the blurring effect of skin reflectance on reflected light through a deconvolution determined by prior knowledge on face material properties. Comparisons to previous techniques for highlight extraction and illumination estimation show the state-of-the-art performance of this approach on a variety of indoor and outdoor scenes.