Towards Digital Refocusing from a Single Photograph

Funck, Wolfram von; Theisel, Holger; Seidel, Hans Peter

doi:10.1109/pg.2007.22

Cited by 28 publications

(18 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [7,13,22,24,26,33,35,52] the authors use a temporally fixed and spatially varying mask in order to estimate the depth, and/or they refocus the out-of-focus part to get an always-in-focus (neat) image. In [16] the authors deal with the question of the optimal trade-off between depth of field and exposure time.…”

Section: Middle: the Image Reconstructed By Direct Deconvolution Rigmentioning

confidence: 99%

A Theory of Optimal Flutter Shutter for Probabilistic Velocity Models

Tendero¹,

Morel²

2016

SIAM J. Imaging Sci.

View full text Add to dashboard Cite

Abstract. Flutter shutter (coded exposure) is a new paradigm for cameras that allows for an arbitrary increase of the exposure time when the relative camera/scene motion is uniform. The photon flux is interrupted according to a flutter shutter code. For arbitrarily severe uniform motion blur a well chosen code guarantees an invertible blur kernel. Yet, when the relative camera/scene velocity is a known constant, a flutter shutter cannot gain more than a 1.17 factor in terms of root mean-squared error compared to the optimal snapshot. In this paper, we prove that this optimality bound can be relaxed under the realistic assumption that a random model for the velocities is available. We give analytical formulae for the optimal flutter shutter code and the optimal snapshots associated with a random velocity distribution. Conversely we also prove formulae that reveal the velocity distribution underlying a given flutter shutter code. 1. Introduction. Digital cameras count at each pixel sensor the number of photons emitted by the observed scene during a time span called the exposure time. The photon count follows a Poisson random variable. Its mean is the ideal (noiseless) pixel value. The difference between this ideal pixel value and the observed sensor photon count is called (shot) noise. The ratio of the mean of the photon count over its standard-deviation is called the signal-to-noise ratio (SNR). In passive imaging systems there is no control over the scene lighting. Thus, the only safe way to increase the SNR is to integrate more photons by increasing the exposure time. Yet, when the scene or the camera moves during the exposure process, the resulting images are degraded by a motion blur. If the support of a motion blur kernel exceeds two pixels, the blur is no longer invertible. This limits the exposure time and therefore the image quality of a snapshot. A setup solving this photography trade-off between noise and blur was proposed in [3,4,6,37,38,39] for uniform camera-scene motions. The authors of these papers attached a flutter shutter to a camera to get an invertible motion blur kernel. A flutter shutter

show abstract

Section: Middle: the Image Reconstructed By Direct Deconvolution Rigmentioning

confidence: 99%

A Theory of Optimal Flutter Shutter for Probabilistic Velocity Models

Tendero¹,

Morel²

2016

SIAM J. Imaging Sci.

View full text Add to dashboard Cite

show abstract

“…An example of this is given in both Figures 1 and 2c. While many have suggested adding depth-of-field to an image after it has been captured, most methods either lack portability, availability, or are too expensive for the average consumer [3], [8], [17]- [20], [22], [23], [29].…”

Section: Depth-of-fieldmentioning

confidence: 99%

“…With sufficient geometric information one can create a variety of interesting effects in images [2], [3], [6], [8], [10], [12], [14], [15], [17]- [23], [28]- [32]. Given a full 3-D reconstruction from an image, one can use conventional rendering tools to re-render the associated scene with new lighting, depth-offield, and an array of other effects.…”

Section: Introductionmentioning

confidence: 99%

“…RELATED WORK Many have suggested methods for introducing depthaware effects into photographs by first estimating depth from a single image, but this is a highly under-determined problem and each such approach makes various assumptions about the scene [2], [3], [6], [21], [28]. Some have used a geometric model for the depth obtained through extensive user interaction (still with various scene assumptions) or a pre-existing model [10], [12], [15].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Interactive Depth-Aware Effects for Stereo Image Editing

Abbott

Morse

2013

2013 International Conference on 3D Vision

View full text Add to dashboard Cite

This paper introduces methods for adding userguided depth-aware effects with minimal user interaction to images captured with a consumer-grade stereo camera. In particular, we present methods for highlighted depth-offield [14], haze, depth-of-field, and image relighting. Unlike many prior methods for adding such effects, we do not assume prior scene models or require extensive user guidance to create such models, nor do we require multiple images of the scene under different conditions. We also do not require specialized camera rigs or other equipment such as light-field camera arrays, active lighting, etc. Instead, we use only an easily portable and affordable consumer-grade stereo camera. The depth is calculated from a stereo image pair using an extended version of PatchMatch Stereo [5] designed to compute not only image disparities but also normals for visible surfaces. We also introduce a pipeline for rendering multiple effects in the order they would occur physically. Each can be added, removed, or adjusted in the pipeline without having to reapply subsequent effects. Individually or in combination, these effects can be used to enhance the sense of depth or structure in images and provide increased artistic control. These effects can be applied to the stereo pair together in a fashion that preserves stereo consistency, or they can be applied to a single image only, thus leveraging the advantages of stereo acquisition even when producing a single photograph.

show abstract

“…Closely related to our work are the other methods for image refocusing: [4] deconvolves a (partially blurry) image with various blur kernel sizes and then chooses for each location the best reconstruction to create a globally sharp image which can be subsequently locally blurred. [5] estimates a depth map by analysing edge profiles, and is able to refocus.…”

Section: Related Workmentioning

confidence: 99%

Automatic foreground-background refocusing

Loktyushin

Harmeling

2011

2011 18th IEEE International Conference on Image Processing

View full text Add to dashboard Cite

A challenging problem in image restoration is to recover an image with a blurry foreground. Such images can easily occur with modern cameras, when the auto-focus aims mistakenly at the background (which will appear sharp) instead of the foreground, where usually the object of interest is. In this paper we propose an automatic procedure that (i) estimates the amount of out-of-focus blur, (ii) segments the image into foreground and background incorporating clues from the blurriness, (iii) recovers the sharp foreground, and finally (iv) blurs the background to refocus the scene. On several real photographs with blurry foreground and sharp background, we demonstrate the effectiveness and limitations of our method.

show abstract

Towards Digital Refocusing from a Single Photograph

Cited by 28 publications

References 30 publications

A Theory of Optimal Flutter Shutter for Probabilistic Velocity Models

A Theory of Optimal Flutter Shutter for Probabilistic Velocity Models

Interactive Depth-Aware Effects for Stereo Image Editing

Automatic foreground-background refocusing

Contact Info

Product

Resources

About