Beyond the Pixel: a Photometrically Calibrated HDR Dataset for Luminance and Color Prediction
Christophe Bolduc     Justine Giroux     Marc Hébert     Claude Demers     Jean-François Lalonde    



description Paper code Code storage Dataset newspaper Poster videocam Video insert_comment BibTeX
Accepted as an Oral presentation in International Conference on Computer Vision (ICCV), 2023!

Abstract

Light plays an important role in human well-being. However, most computer vision tasks treat pixels without considering their relationship to physical luminance. To address this shortcoming, we present the first large-scale photometrically calibrated dataset of high dynamic range 360° panoramas. Our key contribution is the calibration of an existing, uncalibrated HDR Dataset. We do so by accurately capturing RAW bracketed exposures simultaneously with a professional photometric measurement device (chroma meter) for multiple scenes across a variety of lighting conditions. Using the resulting measurements, we establish the calibration coefficients to be applied to the HDR images. The resulting dataset is a rich representation of indoor scenes which displays a wide range of illuminance and color temperature, and varied types of light sources. We exploit the dataset to introduce three novel tasks: where per-pixel luminance, per-pixel temperature and planar illuminance can be predicted from a single input image. Finally, we also capture another smaller calibrated dataset with a commercial 360° camera, to experiment on generalization across cameras. We are optimistic that the release of our datasets and associated code will spark interest in physically accurate light estimation within the community.

Dataset

Laval Photometric Indoor HDR
More information on how to obtain the complete dataset, view a preview and download a sample of 100 panoramas can be found here.
We also provide the train/test/val, split we used in the paper.

Theta Dataset
The Theta Dataset is composed of 74 photometric HDR captured with the Ricoh Theta Z1, an off-the-shelf 360° camera. The photometric HDR as well as the compressed LDR jpeg images are available here.
We also provide the train/test/val, split we used in the paper.

Presentation video


Citation

@inproceedings{bolduc2023beyondthepixel,
	title={Beyond the Pixel: a Photometrically Calibrated {HDR} Dataset for Luminance and Color Prediction},
	author={Bolduc, Christophe and Giroux, Justine and H{\'e}bert, Marc and Demers, Claude and Lalonde, Jean-Fran{\c{c}}ois},
	booktitle={IEEE/CVF International Conference on Computer Vision (ICCV)},
	year={2023}
}

Acknowledgements

This research was supported by Sentinel North, NSERC grant RGPIN 2020-04799, and the Digital Research Alliance Canada. The authors thank Mojtaba Parsaee and Anthony Gagnon for their help with the chroma meter and the Theta Z1 calibration.