Deep outdoor illumination estimation
Citations Over TimeTop 10% of 2017 papers
Abstract
We present a convolutional neural network-based (CNN-based) technique to estimate high-dynamic range outdoor illumination from a single low dynamic range image. To train the CNN, we leverage a large dataset of outdoor panoramas. We fit a low-dimensional physically-based outdoor illumination model to the skies in these panoramas giving us a compact set of parameters (including sun position, atmospheric conditions, and camera parameters). We extract limited field-of-view images from the panoramas, and train a CNN with this large set of input image–output lighting parameter pairs. Given a test image, this network can be used to infer illumination parameters that can, in turn, be used to reconstruct an outdoor illumination environment map. We demonstrate that our approach allows the recovery of plausible illumination conditions and enables photorealistic virtual object insertion from a single image. An extensive evaluation on both the panorama dataset and captured HDR environment maps shows that our technique significantly outperforms previous solutions to this problem.
Related Papers
- → Validation of High Dynamic Range Imaging to Luminance Measurement(2005)32 cited
- → Analysis of tone mapping operators on high dynamic range images(2012)4 cited
- → Perceptually fractural pixel values in rendering high dynamic range images(2010)1 cited
- → High Dynamic Range Display Adopting High Dynamic Range Imaging Technique(2006)4 cited
- Perceptually Fractural Pixel Values in Rendering High Dynamic Range Images(2016)