Glare Aware Photography: 4D Ray Sampling for Reducing Glare Effects of Camera Lenses


Ramesh Raskar,  Amit Agrawal,  Cyrus Wilson and  Ashok Veeraraghavan

Figure 1. Single shot glare reduction. We extract glare components from a single-exposure photo in this high dynamic range scene. Using a 4D analysis of glare inside the camera, we can emphasize or reduce glare. The photo in the middle shows a person standing against a sunlit window. We extract reflection glare generated inside lens and manipulate it to synthesize the result shown on the left. On the right we show the glare-reduced component. Notice that the face is now visible with improved contrast.

Figure 2. Glare reduction without loss of spatial resolution. We captured the photo shown on the left using a randomized pinhole array mask in the camera at f/4. A projector on the top-right corner in the scene (not shown) causes glare. By dividing the captured photo with a calibration photo, the mask effects are removed. Note, however, that the mask makes glare appear as high frequency 2D noise. A glare outlier stencil is obtained by comparing the ratio image with its median filtered output. The glare is reduced by interpolating the pixels in outlier stencil from neighboring pixels. All processing is done in 2D image domain.

Figure 3. Our approach is single-shot and does not require high dynamic range image capture. Thus, photographers can use our camera as a regular handheld camera. Here we show four outdoor examples on glare reduction, where a single photo was captured.


First 'single-shot' approach to classify and reduce glare. We show that glare manifests as an outlier in 4D ray space and use a statistical approach to remove it by 4D sampling without reconstructing a 4D-lightfield.

Glare arises due to multiple scattering of light inside the camera's body and lens optics and reduces image contrast. While previous approaches have analyzed glare in 2D image space, we show that glare is inherently a 4D ray-space phenomenon. By statistically analyzing the ray-space inside a camera, we can classify and remove glare artifacts. In ray-space, glare behaves as high frequency noise and can be reduced by outlier rejection. While such analysis can be performed by capturing the light field inside the camera, it results in the loss of spatial resolution. Unlike light field cameras, we do not need to reversibly encode the spatial structure of the ray-space, leading to simpler designs. We explore masks for uniform and non-uniform ray sampling and show a practical solution to analyze the 4D statistics without significantly compromising image resolution. Although diffuse scattering of the lens introduces 4D low-frequency glare, we can produce useful solutions in a variety of common scenarios. Our approach handles photography looking into the sun and photos taken without a hood, removes the effect of lens smudges and reduces loss of contrast due to camera body reflections. We show various applications in contrast enhancement and glare manipulation.

Paper (Preprint)

Low res pdf (4MB)    High res pdf (27 MB)


SIGGRAPH 2008 Talk

Light Field Datasets for Glare

Light Field Datasets for Refocusing

Building A Handheld Light Field Camera (& History of Integral Imaging)

Related Papers in Computational Photography
SIGGRAPH 2007      Coded aperture and a new theory of light field capture

SIGGRAPH 2006      Coded exposure for motion deblurring

SIGGRAPH 2005      Combining flash/no-flash images using gradient domain algorithms for removing photography artifacts

CVPR 2007              Simultaneous motion deblurring and super-resolution

Copyright 2008 by Mitsubishi Electric Research Labs (MERL), Patents pending

Back to my homepage