Abstract: A system, computer readable medium, and method for dynamically setting a camera's exposure parameters based on face detection are disclosed. When taking a picture or video of a person in front of a bright background, standard exposure algorithms tend to overexpose the background. In one embodiment disclosed herein, a face detection algorithm is run on the current picture or video frame, and the exposure metering region is inset over the detected face. Exposure time, gain, or other exposure parameters may be set based on the pixels within the exposure metering region. In another embodiment, the exposure metering region tracks a moving face according to lag parameters so that the exposure metering region remains substantially over the face. In yet another embodiment, a plurality of faces may be tracked, with the exposure parameters set based on a weighted average of the pixels within the plurality of face-containing exposure metering regions.

Abstract: Several methods and apparatuses for implementing automatic exposure mechanisms for image capturing devices are described. In one embodiment, an orientation detector located in the device determines orientation data for the device. The automatic exposure mechanism projects an orientation vector into an image plane of an image sensor. Next, the automatic exposure mechanism adjusts an initial position of a metering area used for automatic exposure towards a target position based on the projected orientation vector. The automatic exposure mechanism optionally dampens the adjustment of the metering area.

Abstract: Systems, methods, and a computer readable medium for performing an improved blowout prevention process in an image capture device are provided to compensate for occurrences of exposure “blowouts,” i.e., areas in a captured image where pixel brightness exceeds the sensor's dynamic range of capturing capability. In one embodiment, the captured image's histogram may be analyzed to determine if the image is indicative of the presence of exposure blowouts. Once it has been determined that there likely are blowouts in the image, an exposure bias for the image capture device may be set accordingly. Particularly, the exposure value (EV) for the image capture device may be gradually corrected, e.g., by one-eighth of a stop per captured frame, until the image histogram is no longer indicative of blown out regions, at which point the image capture device's exposure value may gradually be corrected back to “normal,” i.e., non-exposure bias compensated, levels.

Abstract: Devices, methods, and computer readable media for performing image orientation detection using image processing techniques are described. In one implementation, an image processing method is disclosed that obtains image data from a first image captured by an image sensor (e.g., from any image capture electronic device). Positional sensor data captured by the device and corresponding to the image data may also be acquired (e.g., through an accelerometer). If the orientation of the device is not reliably discernible from the positional sensor data, the method may attempt to use rotationally invariant character detection metrics to determine the most likely orientation of the image, e.g., by using a decision forest algorithm. Face detection information may be used in conjunction with, or as a substitute for, the character detection data based on one or more priority parameters. Image orientation information may then be included within the image's metadata.

Abstract: Systems, methods, and a computer readable medium for an improved automatic exposure algorithm attempt to classify an image into a particular “scene category,” and, based on the determined scene category, meter the scene according to a generated metering weighting matrix. In one embodiment, the average luminance is calculated for a central exposure metering region of the image and a plurality of peripheral exposure metering regions surrounding the central exposure metering region. Based on comparisons of the average luminance values of the peripheral exposure regions to the average luminance of the central exposure region, a target metering weighting matrix may be generated. In another embodiment, the scene category corresponds to a predetermined metering weighting matrix.

Abstract: Methods and apparatuses for resizing buffered windows. In one aspect of the invention, a method to resize a buffered window on a data processing system includes: determining an estimated size for a window which has a first pixel image of a first size buffered in a first window buffer; allocating a second window buffer which is large enough to buffer the window in the estimated size; and buffering a second pixel image of the window in a second size in the second window buffer. In one example according to this aspect, a portion of a frame buffer is updated to the second pixel image to display the window in the second size. A portion of the second window buffer, storing the data that represents the second pixel image, is clipped to update the corresponding portion of the frame buffer.

Abstract: A system and method for performing Bayer reconstruction of images using a programmable graphics processing unit (GPU) are described herein. A Bayer filtered image in RAW format is uploaded to the GPU, unpacked, and reconstructed. Optionally, the reconstructed image may be transformed into any desired color space and/or displayed by a video card in which the GPU resides. The reconstruction is performed independently on each of the red, blue, and green image fields. The red and blue image fields are reconstructed using first and second interpolation passes in first and second orthogonal directions. Each reconstruction pass preferably employs a two-lobed Lanczos filter. The green image field is interpolated using a single interpolation pass in a direction diagonal to the first and second orthogonal directions, and preferably employs a box filter.

Abstract: A framework for performing graphics animation and compositing operations has a layer tree for interfacing with the application and a render tree for interfacing with a render engine. Layers in the layer tree can be content, windows, views, video, images, text, media, or any other type of object for a user interface of an application. The application commits change to the state of the layers of the layer tree. The application does not need to include explicit code for animating the changes to the layers. Instead, an animation is determined for animating the change in state. The determined animation is explicitly applied to the affected layers in the render tree. A render engine renders from the render tree into a frame buffer for display on the processing device. Those portions of the render tree that have changed relative to prior versions can be tracked to improve resource management.

Abstract: Disclosed is a system for producing images including techniques for reducing the memory and processing power required for such operations. The system provides techniques for programmatically representing a graphics problem. The system further provides techniques for reducing and optimizing graphics problems for rendering with consideration of the system resources, such as the availability of a compatible GPU.

Abstract: Systems, methods, and computer readable media for dynamically adjusting an image capture device's autofocus (AF) operation based, at least in part, on the device's orientation are described. In general, information about an image capture device's orientation may be used to either increase the speed or improve the resolution of autofocus operations. More particularly, orientation information such as that available from an accelerometer may be used to reduce the number of lens positions (points-of-interest) used during an autofocus operation, thereby improving the operation's speed. Alternatively, orientation information may be used to reduce the lens' range of motion while maintaining the number of points-of-interest, thereby improving the operation's resolution.

Abstract: Techniques for reducing the amount of time a camera device requires to automatically adjust focus and/or exposure settings prior to image capture are described. As disclosed, sensor data is used to identify the camera device's orientation. Once known, face detection operations are performed for only that orientation thereby dramatically reducing the amount of processing time required to perform the face detection. The result of the face detection operations may be used to automatically adjust the image capture device's focus and/or exposure.

Abstract: A graphics animation and compositing operations framework has a layer tree for interfacing with the application and a render tree for interfacing with a render engine. Layers in the layer tree can be content, windows, views, video, images, text, media, or other types of objects for an application's user interface. The application commits state changes to the layers of the layer tree. The application does not need to include explicit code for animating the changes to the layers. Instead, an animation is determined for animating the change in state by the framework which can define a set of predetermined animations based on motion, visibility, and transition. The determined animation is explicitly applied to the affected layers in the render tree. A render engine renders from the render tree into a frame buffer. Portions of the render tree changing relative to prior versions can be tracked to improve resource management.

Abstract: Exemplary embodiments of methods and apparatuses to dynamically redistribute computational processes in a system that includes a plurality of processing units are described. The power consumption, the performance, and the power/performance value are determined for various computational processes between a plurality of subsystems where each of the subsystems is capable of performing the computational processes. The computational processes are exemplarily graphics rendering process, image processing process, signal processing process, Bayer decoding process, or video decoding process, which can be performed by a central processing unit, a graphics processing units or a digital signal processing unit. In one embodiment, the distribution of computational processes between capable subsystems is based on a power setting, a performance setting, a dynamic setting or a value setting.

Abstract: Several methods for operating a built-in digital camera of a portable, handheld electronic device are described. In one embodiment, the device receives a user selection (e.g., tap, tap and hold, gesture) of a region displayed on the display screen (e.g., touch sensitive screen). A touch to focus mode may then be initiated in response to the user selection and exposure and focus parameters determined and adjusted. Then, an automatic scene detection mechanism can determine whether a scene has changed. If the scene has changed, then the touch to focus mode ends and a default automatic focus mode initiates. This mode sets a new exposure metering area and focus area prior to determining and adjusting exposure and focus parameters, respectively.

Abstract: A device, method, computer useable medium, and processor programmed to automatically generate tone mapping curves in a digital camera based on image metadata are described. By examining image metadata from a digital camera's sensor, such as the light-product, one can detect sun-lit, high-light, and low-light scenes. Once the light-product value has been calculated for a given image, a tone mapping curve can automatically be generated within the sensor and adjusted appropriately for the scene based on predetermined parameters. Further, it has been determined that independently varying the slopes of the tone mapping curve at the low end (S0) and high end (S1) of the curve results in more visually appealing images. By dynamically and independently selecting S0 and S1 values based on image metadata, more visually pleasing images can be generated.

Abstract: Systems, methods, and computer readable media for removing noise from the luminance (luma) channel in a digital image represented in the YUV color space are described. In general, an element from the luma channel may be selected and a region about that element defined. Using a threshold that is based on the selected luma element's value, similar luma values within the defined region may be identified and combined to provide a substitute value. The substitute value may be blended with the value of the selected element within the image's luma channel. In another implementation, element values from both an image's luma and chroma channels may be used to identify similar luma values.

Abstract: Systems, methods, and computer readable media for performing auto-exposure operations in a digital image capture device. An image capture device's auto-exposure settings may be automatically adjusted by generating and then combining, in a dynamically weighted fashion, two sets of exposure values. One set of exposure values may be based on detected faces while the other set may be based on the image as a whole. Each set of exposure values may include a “measured” value and a “target” value, wherein each of these values may, in turn, be based on one or more (different) aspects of the image. Illustrative digital image capture devices in which the disclosed technology may be used include, but are not limited, to portable digital cameras, mobile telephones, personal digital assistants, mobile music and video devices, and portable computer platforms such as notebook and tablet computers.

Abstract: This disclosure pertains to devices, methods, and computer readable media for performing positional sensor-assisted panoramic photography techniques in handheld personal electronic devices. Generalized steps that may be used to carry out the panoramic photography techniques described herein include, but are not necessarily limited to: 1.) acquiring image data from the electronic device's image sensor; 2.) performing “motion filtering” on the acquired image data, e.g., using information returned from positional sensors of the electronic device to inform the processing of the image data; 3.) performing image registration between adjacent captured images; 4.) performing geometric corrections on captured image data, e.g., due to perspective changes and/or camera rotation about a non-center of perspective (COP) camera point; and 5.) “stitching” the captured images together to create the panoramic scene, e.g., blending the image data in the overlap area between adjacent captured images.

Abstract: An apparatus, method, computer useable medium, and processor programmed to automatically generate tone mapping curves in a digital camera based on image metadata are described. Rather than having a static tone mapping curve for all images, the curve can be varied automatically based on, e.g., the brightness histogram of the image. In one embodiment, a certain percentage of the least bright pixels and a certain percentage of the brightest pixels can be disregarded, while the remaining pixels can be linearly stretched to encompass the original range of brightness values. Based on the distribution of the resultant stretched brightness histogram, slopes for the low end (S0) and high end (S1) of the tone mapping curve can be independently determined, and the tone mapping curve can be automatically generated. The improved, automatically generated tone mapping curve may be able to lift shadows more aggressively and/or increase the dynamic range of the image.

Abstract: Increasing color saturation and contrast in images generally leads to more pleasing images; however, doing so uniformly to all colors in the image can make skin tones appear with an overly red tint. One embodiment of an improved method of skin tone aware color boosting identifies areas of the image which look like skin tones and areas that do not look like skin tones. A blurred “skin tone mask” can then be created over the image. One large boost operation and one small boost operation can be applied to the image. A final version of the image may then be created, applying the pixel values resulting from the small boosting operation to the skin tone regions and applying the pixel values resulting from the large boosting operation to the non-skin tone regions, using the blurred mask to provide a smooth transition between the skin tone and non-skin tone regions.