Efficient spherical high dynamic range imaging for image-based virtual environments

Most high dynamic range (HDR) imaging techniques generate HDR radiance maps from exposure bracketed low dynamic range (LDR) images captured with a stationary camera. We propose a novel general framework for spherical HDR imaging for image-based virtual environments from a moving camera. The framework is composed of three major stages: calibration and alignment, spherical stereo matching and HDR composition. In the first stage, camera poses are found and spherical images are rotationally aligned. In the second stage, disparity maps are calculated with a spherical stereo vision toolkit. In the third stage, spherical images are warped from neighboring views to a target view based on enhanced disparity maps, and a spherical HDR radiance map is obtained from the warped exposure bracket. Our method is efficient because we generate a spherical HDR image for each of the viewpoints of the LDR images. We demonstrate our framework on indoor and outdoor scenes and compare our results with two recent state-of-the-art HDR imaging methods.