Many significant and commercially important uses of modern computer technology relate to images. These include image processing, image analysis and computer vision applications. In computer vision applications, such as, for example, object recognition and optical character recognition, it has been found that a separation of illumination and material aspects of an image can significantly improve the accuracy of computer performance.
The present invention provides a method and system implementing image processing techniques that utilize spatio-spectral information relevant to an image, to perform an operation to accurately and correctly identify and separate illumination and material aspects of the image, to generate an intrinsic image from the image.
In a first exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, organizing information relevant to the image according to a smooth-illumination constraint enforcing a Laplacian filter response such that ∇2ƒ(x, y)=0 and utilizing the smooth-illumination constraint to generate an intrinsic image from the image.
As a feature of the first exemplary embodiment of the present invention, a computer system is provided. The computer system comprises a CPU and a memory storing an image file containing an image. The CPU is arranged and configured to execute a routine to perform the process steps of the method according to the first exemplary embodiment of the present invention. As a further feature of the first exemplary embodiment of the present invention, a computer program product, disposed on a computer readable media is provided. The product includes computer executable process steps operable to control a computer to perform the process steps of the method according to the first exemplary embodiment of the present invention.
In a second exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in an array of pixels, in a computer memory, identifying tokens in the image, as a function of the pixels, modifying the tokens to eliminate blend pixels, organizing the modified tokens according to a constraint and utilizing the constraint to generate an intrinsic image from the image.
As a feature of the second exemplary embodiment of the present invention, a computer system is provided. The computer system comprises a CPU and a memory storing an image file containing an image. The CPU is arranged and configured to execute a routine to perform the process steps of the method according to the second exemplary embodiment of the present invention. As a further feature of the second exemplary embodiment of the present invention, a computer program product, disposed on a computer readable media is provided. The product includes computer executable process steps operable to control a computer to perform the process steps of the method according to the second exemplary embodiment of the present invention.
In a third exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, organizing information relevant to the image according to smooth-illumination constraints, weighting the constraints based upon knowledge of depth discontinuities or occlusions and utilizing the weighted smooth-illumination constraints to generate an intrinsic image from the image.
As a feature of the third exemplary embodiment of the present invention, a computer system is provided. The computer system comprises a CPU and a memory storing an image file containing an image. The CPU is arranged and configured to execute a routine to perform the process steps of the method according to the third exemplary embodiment of the present invention. As a further feature of the third exemplary embodiment of the present invention, a computer program product, disposed on a computer readable media is provided. The product includes computer executable process steps operable to control a computer to perform the process steps of the method according to the third exemplary embodiment of the present invention.
In a fourth exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, organizing information relevant to the image according to smooth-illumination constraints based upon a Laplacian filter response, weighting the constraints as a function of a filter response such as, for example, the filter response: w(x, y)=1/|∇2i(x, y)| and utilizing the smooth-illumination constraints to generate an intrinsic image from the image.
As a feature of the fourth exemplary embodiment of the present invention, a computer system is provided. The computer system comprises a CPU and a memory storing an image file containing an image. The CPU is arranged and configured to execute a routine to perform the process steps of the method according to the fourth exemplary embodiment of the present invention. As a further feature of the fourth exemplary embodiment of the present invention, a computer program product, disposed on a computer readable media is provided. The product includes computer executable process steps operable to control a computer to perform the process steps of the method according to the fourth exemplary embodiment of the present invention.
In a fifth exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, organizing information relevant to the image according to smooth-illumination constraints, detecting smooth-illumination constraint chains as a function of observable image characteristics, adjusting constraints within the detected chain to minimize unwanted color artifact effects in the generation of intrinsic images and utilizing the smooth-illumination constraints to generate an intrinsic image from the image.
As a feature of the fifth exemplary embodiment of the present invention, a computer system is provided. The computer system comprises a CPU and a memory storing an image file containing an image. The CPU is arranged and configured to execute a routine to perform the process steps of the method according to the fifth exemplary embodiment of the present invention. As a further feature of the fifth exemplary embodiment of the present invention, a computer program product, disposed on a computer readable media is provided. The product includes computer executable process steps operable to control a computer to perform the process steps of the method according to the fifth exemplary embodiment of the present invention.
In accordance with yet further embodiments of the present invention, computer systems are provided, which include one or more computers configured (e.g., programmed) to perform the methods described above. In accordance with other embodiments of the present invention, non-transitory computer readable media are provided which have stored thereon computer executable process steps operable to control a computer(s) to implement the embodiments described above. The present invention contemplates a computer readable media as any product that embodies information usable in a computer to execute the methods of the present invention, including instructions implemented as a hardware circuit, for example, as in an integrated circuit chip. The automated, computerized methods can be performed by a digital computer, analog computer, optical sensor, state machine, sequencer, integrated chip or any device or apparatus that can be designed or programmed to carry out the steps of the methods of the present invention.
a is an original image used as an example in the identification of Type C tokens.
b shows Type C token regions in the image of
a-d illustrate a sequence of representations of a Type C token, showing a modification of the Type C token to model the token based upon a token trend prediction, according to a smooth-illumination constraint feature of the present invention.
Referring now to the drawings, and initially to
Alternatively, the CPU 12 can be implemented as a microprocessor embedded in a device such as, for example, the digital camera 14 or a robot. The CPU 12 can also be equipped with a real time operating system for real time operations related to images, in connection with, for example, a robotic operation or an interactive operation with a user.
As shown in
According to a feature of the present invention, in an image process, the CPU 12 operates to analyze and process information, for example, the RGB values of the pixels of an image stored in an image file 18, to achieve various objectives, such as, for example, a correct and accurate identification of illumination and material aspects of the image. The present invention provides a method and system for generation of intrinsic images. The intrinsic images correspond to an original image, for example, an image depicted in an input image file 18. The intrinsic images include, for example, an illumination image, to capture the intensity and color of light incident upon each point on the surfaces depicted in the image, and a material reflectance image, to capture reflectance properties of surfaces depicted in the image (the percentage of each wavelength of light a surface reflects).
Images depicted in some image files 18 are dominated by a single material, for example, a lawn in a sports playing field, or skin in a human face. In such a case, the present invention utilizes the fact of the dominance of a single reflectance region of an image, to improve computer operations relevant to the generation of intrinsic images. To that end,
In step 1000 a selected image file 18 is input to the CPU 12. In step 1002, the CPU 12 executes a routine to detect a dominant image region of a single reflectance. According to a feature of the present invention, any one of several different methods can be implemented to detect the region of single reflectance. For example, step 1002 can be implemented via alternative methods, such as, for example, a log chromaticity clustering method, a large token analysis, a pixel stability analysis or a spectral analysis, as will appear.
In step 1004, the CPU 12 segregates the image as a function of the detected dominant, single reflectance region of the image depicted in the input image file 18. The segregation can be performed by the CPU 12 through execution of any one of several techniques, for example, a constraint/solver model, or a computation based upon pixels of the detected region of single reflectance. In step 1006, the CPU 12 performs post processing, such as, for example, a factoring of the illumination to provide an image illuminated by diffuse illumination, then outputs intrinsic images corresponding to the image of the input image file 18 (step 1008).
In the performance of step 1002 using log chromaticity clustering, the CPU 12 computes an illumination-invariant log chromaticity value for each pixel of the input image file 18. Log chromaticity is a technique for developing a two dimensional illumination invariant chromaticity space. A method and system for separating illumination and reflectance using a log chromaticity space is disclosed in U.S. Pat. No. 7,596,266, which is hereby expressly incorporated by reference. The techniques taught in U.S. Pat. No. 7,596,266 can be used to provide illumination invariant log chromaticity representation values for each pixel of an image.
Logarithmic values of the color band values of the image pixels are plotted on a log-color space graph. The logarithmic values are then projected to a log-chromaticity projection plane oriented as a function of a bi-illuminant dichromatic reflection model (BIDR model), to provide a log chromaticity value for each pixel, as taught in U.S. Pat. No. 7,596,266. The BIDR Model predicts that differing color measurement values fall within a cylinder in RGB space, from a dark end (in shadow) to a bright end (lit end), along a positive slope, when the color change is due to an illumination change forming a shadow over a single material of a scene depicted in the image.
Thus, the RGB values of each pixel in an image file 18 can be mapped by the CPU 12 from the image file value p(n, m, R, G, B) to a log value, then, through a projection to the chromaticity plane, to the corresponding u, v value, as shown in
Accordingly, all of the pixels of the input image file 18 representing the dominant region of single reflectance, for example, pixels corresponding to skin in an image depicting a human face, will have a similar log chromaticity value (within the noise level of the device used to record the image, for example, the camera 14), and be within a single cluster. The CPU 12 can implement any known clustering algorithm, to cluster pixel log chromaticity values. Pixels representing regions of the image other than the skin, for example, eyebrows, lips, the iris of the eye, and so on, will fall outside the single cluster corresponding to skin pixels. Thus, all pixels within the single large cluster will be designated as being in the dominant region of single reflectance.
In the performance of step 1002 using the alternative large token analysis, the CPU 12 initially computes Type C tokens. Referring now to
At step 102, a seed test is begun. The CPU 12 selects a first pixel, i=1, pixel(1, 1) for example (see
If the comparison does not result in approximately equal values (for example, within the noise levels of the recording device for RGB values) for the pixels in the seed, the CPU 12 increments the value of i(step 106), for example, i=2, pixel (1, 2), for a next N×N seed sample, and then tests to determine if i=imax (decision block 108).
If the pixel value is at imax, a value selected as a threshold for deciding to reduce the seed size for improved results, the seed size, N, is reduced (step 110), for example, from N=15 to N=12. In an exemplary embodiment of the present invention, imax can be set at a number of pixels in an image ending at pixel (n, m), as shown in
After reduction of the seed size, the routine returns to step 102, and continues to test for token seeds. An Nstop value (for example, N=2) is also checked in step 110 to determine if the analysis is complete. If the value of N is at Nstop, the CPU 12 has completed a survey of the image pixel arrays and exits the routine.
If the value of i is less than imax, and N is greater than Nstop, the routine returns to step 102, and continues to test for token seeds.
When a good seed (an N×N array with approximately equal pixel values) is found (block 104), the token is grown from the seed. In step 112, the CPU 12 pushes the pixels from the seed onto a queue. All of the pixels in the queue are marked with the current region ID in the region map. The CPU 12 then inquires as to whether the queue is empty (decision block 114). If the queue is not empty, the routine proceeds to step 116.
In step 116, the CPU 12 pops the front pixel off the queue and proceeds to step 118. In step 118, the CPU 12 marks “good” neighbors around the subject pixel, that is neighbors approximately equal in color value to the subject pixel, with the current region ID. All of the marked good neighbors are placed in the region map and also pushed onto the queue. The CPU 12 then returns to the decision block 114. The routine of steps 114, 116, 118 is repeated until the queue is empty. At that time, all of the pixels forming a token in the current region will have been identified and marked in the region map as a Type C token. In the event the pixels comprise intensity histogram representations, the token can be marked as Type CT.
When the queue is empty, the CPU 12 proceeds to step 120. At step 120, the CPU 12 increments the region ID for use with identification of a next token. The CPU 12 then returns to step 106 to repeat the routine in respect of the new current token region.
Upon arrival at N=Nstop, step 110 of the flow chart of
In the token size analysis implementation of step 1002, upon the identification of Type C tokens in the selected input image file 18, the CPU 12 performs an analysis of the identified Type C tokens. Type C tokens that include, for example, skin pixels will be large relative to tokens corresponding to pixels representing other features of the face, for example, eyebrows, lips, the iris of the eye, and so on. Thus, a threshold value for a number of pixels in a token can be set so as to be able to identify relatively large tokens, and thereby establish the region of a dominant, single reflectance. For example, the threshold size can be set as a function of the interocular distance (the distance between the eyes), such as: threshold=(interocular distance in pixels)/7.
In the performance of step 1002 using stability analysis, the CPU 12 operates to characterize the stability of pixels in an input image file 18 and defines a dominant region of single reflectance as a function of the pixel stability measurements. For example, the pixels of identified Type C tokens can be analyzed as to stability, the tokens included in the dominant region if the constituent pixels are stable. In a stability analysis, focus is placed upon how similar a subject pixel is to neighboring pixels. Pixels in a dominant region of an image will be similar to a large number of neighboring pixels. Stability can be measured using several different methods.
For example, second derivative energy can be used to indicate the stability of pixels of an image. In this approach, the CPU 12 calculates a second derivative at each pixel, or a subset of pixels disbursed across the image to cover all illumination conditions of the image depicted in an input image file 18, using a Difference of Gaussians, Laplacian of Gaussian, or similar filter. The second derivative energy for each pixel examined can then be calculated by the CPU 12 as the average of the absolute value of the second derivative in each color band (or the absolute value of the single value in a grayscale image), the sum of squares of the values of the second derivatives in each color band (or the square of the single value in a grayscale image), the maximum squared second derivative value across the color bands (or the square of the single value in a grayscale image), or any similar method. The sigma of the gaussian (defining the size of the second derivative filter) is adjusted to interocular distance when the subject image depicts a face.
Upon the calculation of the second derivative energy for each of the pixels, the CPU 12 analyzes the energy values of the pixels. There is an inverse relationship between second derivative energy and pixel stability, the higher the energy, the less stable the corresponding pixel. In an image having a dominant region of single reflectance, all of the pixels in the dominant region will have a similar low second derivative energy. The CPU 12 can identify all such pixels and thereby identify the dominant region of the image.
In a second method for measuring pixel stability, the CPU 12 utilizes a second derivative relative to a vertical direction. This is an approach based upon a recognition of the fact that certain features of the geometry of a face, for example, eyes, eyebrows, nostrils and lips, tend to be horizontally aligned features of a face. Thus, an image can be evaluated in the vertical direction by convolving an image horizontally with a 1-d gaussian filter of some sigma, and then vertically by a 1-d laplacian of gaussian filter. The sigma values used in each convolution process can be set based upon the interocular distance. The result provides second derivative energy information in the vertical direction, which can be used to detect horizontal stripes formed by eyes, eyebrows and lips, while ignoring vertical features such as vertical shadow boundaries. The second derivative energy in the vertical direction can be converted to a single value per pixel, for a stability analysis similar to the second derivative energy method described above.
According to a third method, the CPU 12 applies a bilateral filter to the image. A bilateral filter is an edge preserving blurring filter. The filter changes each pixel value to a weighted average of pixels surrounding a subject pixel. The weights in the weighted average correspond to how similar the surrounding pixels are to the subject pixel both in terms of spectral properties and spatial properties. Thus, a nearby pixel of similar color will have a high weight, while a distant pixel of dissimilar color will have a low weight.
While the weighted average is the typically used output of a known bilateral filter, according to a feature of the present invention, the sum of the weights that go into the weighted average is used to measure pixel stability. The higher the weight, the more similar a pixel is to more of nearby neighbors, and thus, the more stable it is. Each pixel in a dominant region of single reflectance will have a sum of weights for the surrounding pixels which is higher than the pixels representing other areas of the image. The size of the bilateral filter and the threshold for the sum of weights can be adjusted according to the interocular distance. As in the previous methods, this factor is evaluated by the CPU 12 to identify the dominant region in the image.
Stability of pixels can also be evaluated relative to local variance in pixel values. The CPU 12 calculates and evaluates the variance in red, green and blue values in a region surrounding each pixel (or variance in the single value for a grayscale image). The size of the region can be based upon interocular distance in an image depicting a face. The higher the variance, the less stable the pixel. Thus, the evaluation can be based upon the inverse relationship between stability and variance to identify a dominant region.
Each of the above methods for detecting pixel stability produces a value corresponding to the stability of each pixel. The stability values can be used to detect a dominant region by placing a threshold on the values at a preselected value determined empirically: for instance, all pixels which have a sufficiently low variance can be considered stable. The pixel stabilities can be optionally blurred prior to thresholding. Additionally, the resulting region after thresholding can be optionally eroded to remove isolated points and/or dilated to fill in holes.
According to a further exemplary embodiment of the present invention, the performance of step 1002 can be implemented using a spectral analysis. For example, if it is known or expected that the dominant region is a particular color, for example, green for grass in the playing filed of a stadium, then all pixels that are sufficiently similar to the expected color of the dominant region are placed in the region.
Once a dominant region of single reflectance is identified (step 1002), the CPU 12 proceeds to step 1004 to segregate the image into illumination and material aspects of the image. If, for example, the log chromaticity method was implemented to identify the region of single reflectance, the CPU 12 can identify the illumination aspects of the image via the relationship of I=ML, or I=RL, wherein I is the intensity of an image reflectance and L the illumination value. In the situation of an image with a single dominant region of same reflectance, the CPU 12 can assume any color for all of the pixels within the single large region identified by the CPU 12, for example, a flesh tone, and then solve the I=ML equation for the illumination value of each pixel, using the assumed material color and the recorded value for I, for each pixel of the single dominant region. The result is an illumination map corresponding to the pixels of the dominant region of single reflectance.
Any known in-filling or in-painting technique can be used to expand the illumination found for the dominant region of single reflectance, to pixels representing other areas of the image, for example, the eyes of an image depicting a human face. For example, a thin plate spline can be fit to the calculated illumination values and used to project the values for the illumination in other areas. For a more detailed description of in-filling or in-painting techniques, reference is made to “Inpainting and the Fundamental Problem of Image Processing” by Jianhong (Jackie) Shen, SIAM News, Volume 36, Number 5, June 2003.
Alternatively, a constraint/solver model can be implemented to segregate the image (step 1004), for example, when, a tokenization was performed to identify the dominant region of single reflectance. All of the Type C tokens identified as being in the region of single reflectance are constrained in a matrix equation, to be of a same material. All adjacent Type C tokens of the image of the image file 18 are also constrained in a matrix equation to be in a smooth-illumination constraint. A smooth-illumination constraint can be used to determine the illumination in regions of the image including regions of the image that are outside the dominant region. In a smooth-illumination constraint, the constraint is based upon the average illumination of the pixels near a shared boundary between adjacent Type C tokens. This constrains the illumination field to be somewhat smooth, as opposed to piecewise constant throughout a token.
According to a feature of the present invention, the constraint/solver model can be utilized to segregate illumination and material in an image other than one wherein a dominant region of single reflectance is present. Generally, a same material constraint, as taught, for example, in U.S. Patent Publication No. US 2010/0142825, utilizes Type C tokens and identified regions of uniform reflectance, such as the dominant region of single reflectance. The dominant region of single reflectance comprises a Type B token. The constraining relationship of a same material constraint is that all Type C tokens that are part of the same Type B token are constrained to be of the same material.
As taught in U.S. Patent Publication No. US 2010/0142825, the same material constraint enforces the definition of a Type B token, that is, a connected image region comprising contiguous pixels that represent a region of the image encompassing a single material in the scene (uniform reflectance), though not necessarily the maximal region corresponding to that material. Thus, all Type C tokens that lie within the same Type B token are by the definition imposed upon Type B tokens, of the same material, though not necessarily of the same illumination. The Type C tokens are therefore constrained to correspond to observed differences in appearance that are caused by varying illumination.
Since: ma=is−la, mb=ib−lb, and mc=ic−lc, these mathematical relationships can be expressed, in a same material constraint, as (1)la+(−1)lb+(0)l=(ia−ib), (1)la+(0)lb+(−1)lc=(ia−ic) and (0)la+(1)lb+(−1)lc=(ib−ic).
Thus, in the matrix equation of
According to a feature of the present invention, the smooth-illumination constraint can be implemented as a Laplacian-smooth illumination constraint. In a Laplacian-smooth illumination constraint, the concept of smooth is defined in a manner such that an illumination image has a local response to a Laplacian filter that is zero. The Laplacian is computed from the sum of the second derivatives in the x and y directions, as follows:
∇2ƒ(x,y)=∂2ƒ(x,y)/∂x2+∂2·f(x,y)/∂y2 (1)
The Laplacian constraint, ∇2ƒ(x, y)=0, is met when the local variation of the two dimensional function ƒ(x, y) is planar, since the second derivative of a plane is zero.
In the image formulation model of I=RL, the log domain is once again used such that i=r+l (where x≈log (X) for xε{i, r, l}) Thus, the zero-Laplacian constraint is applied to the log-illumination image, as follows:
∇2l=0 (2)
∇2(i−r)=0 (3)
∇2r=∇2i (4)
where the unknowns are the log-reflectance values r, and the right hand side is computed from known, observed log-image data for i, as for example the recorded values stored in an image file 18.
Approximating the second derivative to represent ∇2ƒ(x, y) with discrete centered differences provides the following common Laplacian filtering kernel:
Convolving the above filtering kernel with ƒ(x, y) computes ƒ(x, y−1)+ƒ(x, y+1)+ƒ(x−1, y)+ƒ(x+1, y)−4ƒ(x, y) at each pixel p(n, m) wherein the n, m values are represented by x, y coordinates. Thus, according to the relationship (4), set forth above, a linear system of constraints can be developed on the unknown value r by applying the filtering kernel of (5) to the log-image data to generate a right-hand side of the linear system of constraints and building a left-hand matrix enforcing the same linear filtering relationship on the unknown values, as follows:
The sparse, band-diagonal constraint matrix A has as many rows and columns as there are pixels in the image. The number of non-zero elements will be equal to fiver times the number of pixels.
According to another feature of the present invention, a Laplacian-of Gaussian filtering strategy can be implemented for additional smoothing and/or noise reduction. In such a case, the Laplacian filtering kernel (representation (5), above) is convolved with a two dimensional Gaussian kernel to incorporate information from a larger neighborhood around each pixel. This yields more non-zero entries in A.
For multi-band imagery, such as the RGB data of our example, the constraints represented by (6) and (7) above, are applied independently to each band i(k)(x, y) where, for example, kε{0, 1, 2, . . . } or kε{R, G, B}. In a multi-band image, the right hand side of (7) will have as many columns as there are bands in the multi-band image, for example, three columns in an RGB image. Correspondingly, the solution r will have the same number of columns. The structure and elements of the constraint matrix A remains the same for all bands.
It should also be noted that while the above described Laplacian-smooth illumination constraint was described relative to image pixels, the constraint can be implemented at the Type C token level. In such a case, the filtering kernel entries on each row of A are accumulated for each token in an image of an input image file 18, to provide a new matrix A′. The matrix A′ has as many rows as there are pixels in the image, but only as many columns as there are Type C tokens in the image. The number of tokens will be far fewer than the number of pixels to thereby improve the computational efficiency of CPU 12 operation. The matrix A′ is represented as follows:
where i represents the linear index of pixel (xi, yi) and Tj represents the set of pixels in token j.
In practice, tokens may contain blend pixels. Blend pixels are pixels between two differently colored regions of an image. If the colors between the two regions are plotted in RGB space, there is a linear transition between the colors, with each blend pixel, moving from one region to the next, being a weighted average of the colors of the two regions. Thus, each blend pixel does not represent a true color of the scene depicted in the image, and can cause artifacts in the results of an image process. In such cases, when tokens include blend pixels, there may be a discrepancy in the constraint relationship represented by (7) between the right-hand side filter responses ∇2i computed on the original image data and the left-hand side constraints encoded in the structure A′. If the illumination l is computed as the residual i−r, the result may include unwanted color artifacts in the computed illumination image.
According to a feature of the present invention, the unwanted color artifact effect is mitigated by implementing modifications to the image file 18 before a Laplacian filtering step, and using the modified image for the right-hand side of (7). One modification includes the step of replacing each pixel in the log-image i by the average image value of the token to which each respective pixel is a constituent pixel, as follows:
a-d illustrate another modification to an input image file 18 to mitigate an unwanted color artifact effect.
a shows a Type C token. In
As shown in
Next, a distance transform is performed on the interior pixels. For every distance of 1 or less for which there is no model based upon a thin-plate spline, the pixel value in the window is specified by the value of the pixel that lead to the low distance. Thus, any interior pixels for which there was not a thin-plate spline, are modeled by themselves, and any perimeter pixels adjacent to an interior pixel is modeled by the interior pixel. Finally, the values of perimeter pixels are projected out one pixel, such that any pixel within one pixel of the token is modeled by the nearest token pixel and any pixel on the perimeter of the token that is not neat a thin-plate spline or an interior pixel is modeled by itself.
Moreover, if a token has eight or fewer pixels it cannot have interior pixels. Accordingly, in such cases, the modeling process is not performed. Instead, the CPU 12 executes a code for querying a model value to loop through the eight or fewer pixels of the respective token, looking for a pixel spatially closest to the pixel to be modeled, returning that value for the pixels of the token.
According to another feature of the present invention, when solving for an optimal reflectance r according to the above described constraints, the constraints can be individually weighted to emphasize or de-emphasize each constraint, as desired by a user. The weights assigned to various constraints can be specified both relative to each other and relative to constraints from other sources, for example, constraints in a same material constraint matrix are weighted relative to one another, or same material constraints are weighted relative to smooth illumination constraints, and so on. In general, Laplacian-smooth illumination constraints are assigned much less weight, relative to other constraint types (by an order of magnitude or two).
In addition, other sources of information are used to weight individual smooth-illumination constraints relative to one another. For example, when there is knowledge of depth discontinuities or occlusions, from a source such as stereo information or user input, there is no reason to enforce a smooth-illumination constraint because at a depth discontinuity it is probable that spatially-separated surfaces are lit by different illumination conditions. Thus, the constraints (i.e. rows in A) corresponding to the pixels of an image located at the discontinuity can be down-weighted substantially or possibly removed entirely.
Another possibility, according to this feature of the present invention, is to down weight the constraints as a function of a filter response. For example, the magnitude of the actual filter response, |∇2i(x, y)|, can be used to weight the constraint at pixel (x, y). Since the Laplacian filter responds more strongly at sharp boundaries, and the underlying assumption is that illumination transitions are smooth, constraints at pixels with stronger filter responses are subject to a down-weight adjustment. A weighting scheme in such cases can be implemented as the simple inverse of the filter response, as follows: w(x, y)=1/|∇2i(x, y)|, or alternatively, one over the square of the Laplacian, one over the square root of the Laplacian or one over the magnitude of the gradient at the image location.
According to yet another feature of the present invention, smooth-illumination constraint chains that are impossible under reasonable illumination conditions are detected and broken. Smooth-illumination constraints can be implemented between adjacent Type C tokens. Thus, in an example of tokens a, b and c, there is a constraint between tokens a and b, and also between tokens b an c, and so on, to form a chain of constraint relationships.
Smooth-illumination constraint chains can cause unwanted color artifact effects in the generation of intrinsic images. A usual reason is because a Type C token representing a lit foreground object borders a Type C token representing a shadowed background (or vice versa). Such a circumstance can be so problematical that the foreground is brightened in an intrinsic image beyond what is reasonable. According to a feature of the present invention, a chain of tokens causing an unwanted color artifact can be detected by searching the chain as a function of certain observable image characteristics, even without depth information, or user input.
For example, for each Type C token identified in an image, an estimate can be made of the lower bound for the brightness of the material color of the respective token. If a Type C token is part of a Type B token, then the material color brightness for the token must be at least as bright as the brightest part of the respective Type B token in the original image. If the Type C token is not part of an identified Type B token, then it must be at least as bright as the respective token appears in the original image. Either of those values provides a basis for an estimate of a lower bound for maximum brightness of a token. Moreover, due to the constraint between tokens a and b, and the constraint between tokens b and c, an analysis can be made relative to a reasonable relationship between tokens a and c.
According to a feature of the present invention, the image characteristics of a lower bound for material brightness, and a reasonable relationship between tokens a and c, are used to determine whether a chain is reasonable. A Type C token is selected, and the minimum brightness for the selected token is estimated (one of the brightest part of a Type B token to which the selected token belongs, or the brightness of the Type C token in the original image). Next, results for smooth-illumination constraints along a chain of tokens from the selected token, are determined.
If the resulting brightness according to the chain of constraints is significantly higher than the brightest value possible for a token in the chain, for example, 1.5 times an estimated brightness for the token, the chain is unreasonable. Generally, if at any point in the execution of a smooth-illumination constraint, a predicted value for a token is more than 1.5 times the brightest value in the original image, the constraint is impossible. The weight for each constraint marked as impossible is significantly weakened when solving for optimum values for an intrinsic image, as will be described. Each impossible constraint is weakened rather than removed to maintain continuity in the analysis.
Additionally, a check is made as to how much darker a last token in a chain is than the first token of the chain. If the last token is more than 40 times darker than the first token, that is another indication that the chain is unreasonable. This is based upon knowledge that material variation is always less than 40:1. The last constraint of that chain is also marked as an impossible constraint.
A standard least squares solver can be executed by the CPU 12 to solve for optimized values for the unknowns in the matrix equations of
In the preceding specification, the invention has been described with reference to specific exemplary embodiments and examples thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention as set forth in the claims that follow. The specification and drawings are accordingly to be regarded in an illustrative manner rather than a restrictive sense.
Number | Name | Date | Kind |
---|---|---|---|
6611627 | LaRossa et al. | Aug 2003 | B1 |
8139850 | Maxwell et al. | Mar 2012 | B2 |
8139867 | Maxwell et al. | Mar 2012 | B2 |
8249342 | Friedhoff et al. | Aug 2012 | B1 |
8260050 | Stein et al. | Sep 2012 | B2 |
8385655 | Smith et al. | Feb 2013 | B2 |
8428352 | Smith | Apr 2013 | B1 |
8483480 | Smith et al. | Jul 2013 | B2 |
8553979 | Smith et al. | Oct 2013 | B2 |
8559714 | Smith | Oct 2013 | B2 |
20070258645 | Gokturk | Nov 2007 | A1 |
20090161992 | Taguchi et al. | Jun 2009 | A1 |
20090245651 | Friedhoff et al. | Oct 2009 | A1 |
20100061628 | Yamada | Mar 2010 | A1 |
20100142825 | Maxwell et al. | Jun 2010 | A1 |
20100142846 | Tolliver et al. | Jun 2010 | A1 |
20100157072 | Luo et al. | Jun 2010 | A1 |
20120114263 | Stein et al. | May 2012 | A1 |
20130071033 | Stein et al. | Mar 2013 | A1 |
20130129208 | Friedhoff et al. | May 2013 | A1 |
20130136348 | Smith et al. | May 2013 | A1 |
20130321419 | Stein | Dec 2013 | A1 |
20140050399 | Lalonde et al. | Feb 2014 | A1 |
20140072210 | Stein et al. | Mar 2014 | A1 |
Number | Date | Country | |
---|---|---|---|
20120114263 A1 | May 2012 | US |