Abstract: An angled optical pattern is decoded. To decode an optical pattern imaged at an angle, an area of interest of an image is received. A start line and an end line of the optical pattern are estimated. Corners of the optical pattern are localized. A homography is calculated based on the corners. And a scanline of the optical pattern is rectified based on the homography.

Abstract: An image processing apparatus capable of preventing recurrence of a jam at the time of document reading is provided. The image processing apparatus includes a feeder, a reader, a storage device, and a processor. The processor causes the image processing apparatus to function as: a reading unit configured to cause the reader to read a document by a reading method among a plurality of reading methods including a conveyance-based reading method using the feeder; a storage control unit configured to, in a case where reading of a document with the reader by the conveyance-based reading method has failed, store information on the document in the storage device; and a notification unit configured to, in a case where an instruction to read again the document by the conveyance-based reading method is issued, issue, in accordance with the information, a notification that prompts to change the reading method set for the document.

Abstract: An image processing apparatus, includes a memory; and a processor coupled to the memory and the processor configured to: identify a first recognition error, the first recognition error being an error between ground truth data and a first recognition result obtained by inputting a first feature of image data into a first image recognition model, generate a second feature obtained by adding noise to the first feature of the image data, identify a second recognition error, the second recognition error being an error between the first recognition result and a recognition result obtained by inputting the second feature into a second image recognition model, and execute training of the first image recognition model and the second image recognition model based on the first recognition error and the second recognition error.

Abstract: Systems and techniques for modifying a subsection of uploaded media are presented. An instruction component receives a media file and a media enhancement instruction that includes enhancement data and media interval data for a first segment of the media file. A processing component modifies the first segment of the media file associated with the media interval data based on the enhancement data to generate an edited first segment of the media file. A finalization component generates an edited version of the media file that includes the edited first segment of the media file and at least a second segment of the media file that is not modified based on the enhancement data.

Abstract: Embodiments of the present disclosure relate to user interfaces and systems that may enable dynamic and interactive access of, investigation of, and analysis of data objects stored in one or more databases. The data objects may be accessed from the one or more databases, and presented in multiple related portions of a display. In particular, the system provides a time-based visualization of data objects (and/or properties associated with the data objects) to a user such that the user may, for example, determine connections between various data objects, observe flows of information among data objects, and/or investigate related data objects.

Abstract: A system, method and computer program product for use in editing digital documents with handwriting input to a computing device are provided. The computing device is connected to an input device in the form of an input surface. A user is able to provide input by applying pressure to or gesturing above the input surface using either his or her finger or an instrument such as a stylus or pen. The present system and method monitors the input strokes. The computing device further has a processor and at least one application for recognizing the handwriting input under control of the processor. The at least one system application is configured to cause display of, on a display interface of a computing device, first digital ink in accordance with document content, detect handwriting input as one or more editing gestures with respect to the document content, and, cause display of, on the display interface, second digital ink in accordance with handwriting input.

Abstract: Users of a database management engine may generate fillable digital documents by mapping interface elements onto form documents. When a user maps interface elements onto a form document, the user may accidentally overlap two or more interface elements. To rectify this, the database management engine may modify the position of one of interface elements based on a set of positioning rules. In addition, the database management engine may identify and suggest mappings to users based on similar documents that have been previously mapped. The database management engine identifies similar documents using information about the document, the user, and the mapping itself. The mapping associated with the most similar document may be provided to the user as a suggested mapping. The database management engine converts the form document and finalized mapping into a fillable digital document. The fillable digital document is sent to recipients, who complete the fillable digital document.

Abstract: Methods, systems and computer-readable storage media for scanning a document to detect a source of the document. The document can be scanned to obtain content from the document. An automatic search of an electronic network can be then performed to determine if there is an exact match or a partial match between the content of the document and the results of search automatic search. If an exact match or a partial match is generated as a result of automatically searching the electronic network, a user can be presented with various selection options via a GUI for retrieving, viewing and printing search results of the automatic search.

Abstract: A method, a structure, and a computer system for decomposing speech. The exemplary embodiments may include one or more encoders for generating one or more encodings of a speech input comprising rhythm information, pitch information, timbre information, and content information, and a decoder for decoding the one or more encodings.

Abstract: A document scanner with the capability of landing a document by determining its location on the scanning bed, using intelligent methodologies for framing the document to determine the edges/boundaries of the document, processing the image using various techniques and methodologies to speed up the processing, so that the document feed track does not have to be slowed, and a unique binarization of the image to efficiently create different renderings of the scanned image.

Abstract: There is provided an image reading apparatus including a reading unit configured to sequentially read respective pages of a target original consisting of a plurality of pages to generate read images, and a determining unit configured to determine whether there is a missing page in the read images. The determining unit is configured to determine whether there is a missing page in the read images by applying, to a learned model, determination data based on a pair of consecutively-generated read images of the target original. The learned model is a model that has been derived through machine teaming using input data based on image pairs of consecutive pages and image pairs of non-consecutive pages of the original for learning, and training data that indicates whether each image pair is a pair of consecutive pages.

Abstract: The present invention concerns a novel method to efficiently score documents (texts, images, audios, videos, and any other information file) by using a machine-learned ranking function modeled by an additive ensemble of regression trees. A main contribution is a new representation of the tree ensemble based on bitvectors, where the tree traversal, aimed to detect the leaves that contribute to the final scoring of a document, is performed through efficient logical bitwise operations. In addition, the traversal is not performed one tree after another, as one would expect, but it is interleaved, feature by feature, over the whole tree ensemble. Tests conducted on publicly available LtR datasets confirm unprecedented speedups (up to 6.5×) over the best state-of-the-art methods.

Abstract: Embodiments relate to an intelligent computer platform to create a document specific error correction model for amending OCR values. An image of a document is received and OCR is applied to the received image. Text is extracted from at least one static content field and the extracted text is compared to stored text from known static content. Responsive to a deviation identified in the comparison, a document specific error correction model is created and leveraged to correct OCR output. The model generates one or more variants for the dynamic content field associated with the static content field having the identified deviation. The generated variant(s) is subject to processing and one of the variants is selected as amended document content. A new document version is created from the amendment, the new document version including corrected OCR output.

Abstract: Users of a database management engine may generate fillable digital documents by mapping interface elements onto form documents. When a user maps interface elements onto a form document, the user may accidentally overlap two or more interface elements. To rectify this, the database management engine may modify the position of one of interface elements based on a set of positioning rules. In addition, the database management engine may identify and suggest mappings to users based on similar documents that have been previously mapped. The database management engine identifies similar documents using information about the document, the user, and the mapping itself. The mapping associated with the most similar document may be provided to the user as a suggested mapping. The database management engine converts the form document and finalized mapping into a fillable digital document. The fillable digital document is sent to recipients, who complete the fillable digital document.

Abstract: A non-transitory computer readable medium storing a program that causes a computer to execute a process, the process including obtaining text data, generating first index information indicating appearance positions in the text data for each of a plurality of characters or words obtained based on lexical analysis of the text data, generating second index information, the second index information being index information in which the appearance positions in the text data are aggregated for each character or word, specifying a data range in the first index information, to be referred in a pattern match search by using the second index information, and performing encoding for the text data based on the pattern match search by using the data range in the first index information.

Abstract: Provided are a method and system for effecting a payment transaction. The method capturing a digital image of a price identifier and a merchant identifier at a client device. Generating payment request data by a payer application at the client device comprising data corresponding to the price and merchant identifiers contained in the digital image. Forwarding the payment request data from the payer application to a payment server. Initiating by the payment server a payment transaction from a payer account to a recipient account in response to receipt of the payment request data. Forwarding notification data to the payer application indicating the status of the payment transaction from the payment server.

Abstract: Provided is an excellent robot device capable of preferably detecting difference between dirt and a scratch on a lens of a camera and difference between dirt and a scratch on a hand. A robot device detects a site in which there is the dirt or the scratch using an image of the hand taken by a camera as a reference image. Further, this determines whether the detected dirt or scratch is due to the lens of the camera or the hand by moving the hand. The robot device performs cleaning work assuming that the dirt is detected, and then this detects the difference between the dirt and the scratch depending on whether the dirt is removed.

Abstract: Techniques are provided to implement the display of synchronized sensor data and captured video signal. A mobile computing device receives a set of data points from a multi-sensor device. The set of data points may be assigned a timestamp value, obtained from an internal clock within the mobile computing device. Sets of data points and their assigned timestamp values are aggregated into an aggregated sensor dataset that represents the sets of data points over a particular period of time. Concurrently to receiving the sensor data, video frames are received. The video frames with an assigned timestamp value are stored in a video frame dataset that represents video frames over a particular period of time. The aggregated sensor dataset and the video frame dataset are synchronized and combined into a single combined video dataset based upon the assigned timestamp values of the sets of data points and the video frames.

Abstract: When a reading error occurs while reading out image data saved in a hard disk, input data which is saved in a RAM is read out and image data is regenerated from the read input data. This enables reprinting by an image formation apparatus alone without requiring a user to re-input image data even when the reading of data occurs due to abnormality of the hard disk.

Abstract: An example of an apparatus including an input device to receive details about an event is provided. The apparatus further includes a camera to capture image data at the event. The apparatus also includes a memory storage unit to store the image data captured by the camera. In addition, the apparatus includes a reminder generator to generate a plurality of reminders during the event based on the details. Each reminder of the plurality of reminders requests an image from the camera. Furthermore, the apparatus includes a processor to index the image data in a database on the memory storage unit. The database is to be used to generate a gallery about the event.

Abstract: Systems and methods for automatic processing of receipts to capture data from the receipts are presented. Upon receiving an image of a receipt, an optical character recognition (OCR) of the receipt content embodied in the image is executed. The OCR of the receipt content results in machine-encoded text content of the receipt content embodied in the image. Tokens are generated from the machine-encoded text content and data groups are constructed according to horizontal lines of generated tokens. Potential product items are identified for at least some of the constructed data groups and the potential product items are evaluated for the at least some constructed data groups. The evaluation of the potential product items for the at least some constructed data groups identifies receipt data, such as product items and vendor information, associated with the least some constructed data groups. The identified receipt data is captured and stored with the image of the receipt in a data store.

Abstract: According to one embodiment, a camera input interface includes a first channel processing circuit, a second channel processing circuit, a selector, and a comparator. The first channel processing circuit is configured to perform image processing to image data from the first input source. The second channel processing circuit is configured to perform image processing to image data from the first input source or the second input source. The selector is provided on an input side of the second channel processing circuit and configured to be switched to either one of the first input source and the second input source. The comparator is configured to compare a first output result output from the first channel processing circuit and a second output result output from the second channel processing circuit with each other.

Abstract: A digital image imported into a target application that provides a functionality with respect to digital images may be rendered to obtain a rendered digital image. The digital image may be parsed to obtain asset metadata characterizing the digital image. Position information identifying a position of an image element within the digital image may be used to overlay a virtual layer onto a rendition image element of the rendered digital image corresponding to the image element, within a graphical user interface (GUI) of the target application. In this way, the functionality may be provided within the GUI and with respect to the virtual layer, to thereby provide the appearance of providing the functionality with respect to the rendition image element.

Abstract: An image processing apparatus includes a deterioration degree estimation unit, a statistical information calculation unit, a correction coefficient calculation unit, and a contrast correction unit. The deterioration degree estimation unit estimates a deterioration degree of each local region of image data. The statistical information calculation unit calculates at least a luminance of a high luminance portion in a luminance distribution of the image data from statistical information of each local region of the image data. The correction coefficient calculation unit calculates a correction coefficient to perform contrast correction. The contrast correction unit performs the contrast correction for each local region of the image data on the basis of the correction coefficient calculated by the correction coefficient calculation unit.

Abstract: An image forming device includes: an image forming unit that forms an image on a transfer medium in accordance with a JOB; and a controller that controls the image forming unit. The controller is capable of acquiring a read image obtained by reading the image on the transfer medium, and determining failure/no-failure in the image on the transfer medium in accordance with the read image. When the image is determined to have a failure in accordance with the determination, the controller saves, for each JOB, original image information related to at least a failure-detected page in which the failure has been detected in a state in which the original image information is associated with output history information, and makes at least the failure-detected page re-printable in accordance with the associated original image information.

Abstract: In one aspect, an example method includes (i) capturing, by a computing system, via a camera of the computing system, video content; (ii) encoding, by the computing system, the captured video content to generate a video stream; and (iii) while generating the video stream: (a) making, by the computing system, a determination that each condition in a condition set is satisfied, wherein the condition set includes a condition relating to a change in a data rate of the video stream; and (b) responsive to making the determination, outputting, by the computing system, via a user interface of the computing system, an indication that the computing system is unsteady.

Abstract: The present disclosure discloses a photo processing method and an apparatus for grouping photos into photo albums based on facial recognition results. The method includes: performing face detection on multiple photos, to obtain a face image feature set, each face image feature in the face image feature set corresponding to one of the multiple photos; determining a face-level similarity for each pair of face image features in the face image feature set; determining a photo-level similarity between each pair of photos in the multiple photos in accordance with their associated face-level similarities; generating a photo set for each target photo in the multiple photos, wherein any photo-level similarity between the target photo and another photo in the photo set exceeds a predefined photo-level threshold; and generating a label for each photo set using photographing location and photographing time information associated with the photos in the photo set.

Abstract: A wearable computing device for detecting photography includes an outer casing configured to be worn by a user. The wearable computing device also includes a device camera coupled to the outer casing. The device camera is designed to detect image data corresponding to a person holding a remote camera in an environment of the wearable computing device. The wearable computing device also includes a mobile processor coupled to the device camera. The mobile processor is designed to determine that a photograph will be taken based on the image data corresponding to the person holding the remote camera. The wearable computing device also includes an output device coupled to the mobile processor. The output device is designed to output data indicating that the photograph of the wearable computing device will be taken.

Abstract: A graphical user interface that is embodied in a computer-readable medium for execution on a computer, and configured for processing of an image that is displayed on a screen. The graphical user interface includes an original version of a control element, and a blurred version of the control element. The blurred version of the control element is superimposed over the image on the screen, and the original version of the control element is superimposed over the blurred version of the control element on the screen.

Abstract: Systems and methods for automatic processing of receipts to capture data from the receipts are presented. Upon receiving an image of a receipt, an optical character recognition (OCR) of the receipt content embodied in the image is executed. The OCR of the receipt content results in machine-encoded text content of the receipt content embodied in the image. Tokens are generated from the machine-encoded text content and data groups are constructed according to horizontal lines of generated tokens. Potential product items are identified for at least some of the constructed data groups and the potential product items are evaluated for the at least some constructed data groups. The evaluation of the potential product items for the at least some constructed data groups identifies receipt data, such as product items and vendor information, associated with the least some constructed data groups. The identified receipt data is captured and stored with the image of the receipt in a data store.

Abstract: A method for area extraction includes obtaining an area position of a first information area in a document image, determining a second information area according to the area position of the first information area, and segmenting the second information area to obtain at least one character zone.

Abstract: Techniques are provided to implement the display of synchronized sensor data and captured video signal. A mobile computing device receives a set of data points from a multi-sensor device. The set of data points may be assigned a timestamp value, obtained from an internal clock within the mobile computing device. Sets of data points and their assigned timestamp values are aggregated into an aggregated sensor dataset that represents the sets of data points over a particular period of time. Concurrent to receiving the sensor data, video frames are received. The video frames with an assigned timestamp value are stored in a video frame dataset that represents video frames over a particular period of time. The aggregated sensor dataset and the video frame dataset are synchronized and combined into a single combined video dataset based upon the assigned timestamp values of the sets of data points and the video frames.

Abstract: A facial recognition search system identifies one or more likely names (or other personal identifiers) corresponding to the facial image(s) in a query as follows. After receiving the visual query with one or more facial images, the system identifies images that potentially match the respective facial image in accordance with visual similarity criteria. Then one or more persons associated with the potential images are identified. For each identified person, person-specific data comprising metrics of social connectivity to the requester are retrieved from a plurality of applications such as communications applications, social networking applications, calendar applications, and collaborative applications. An ordered list of persons is then generated by ranking the identified persons in accordance with at least metrics of visual similarity between the respective facial image and the potential image matches and with the social connection metrics.

Abstract: A computer receives a query including a plain text word. The computer determines whether at least one performance optimization parameter is configured. The computer identifies syllables within the plain text word that have a high probability of spelling errors. The computer selects each character in the syllables identified. The computer identifies a group of characters from a confusion matrix that are commonly confused with the character selected. For each character selected, the computer generates a set of characters including characters from the confusion matrix. The computer determines probabilities of omitting the selected characters and associates tags with one or more sets of characters based on the determined probabilities. The computer generates a regular expression by concatenating the sets of characters. The computer searches for a spelling of the plain text word that omits at least one character in the regular expression that is associated with a tag.

Abstract: The disclosure is related to a method and a system for removing background of a video. One of the objectives of the method and system is to extract a motion image object from a video. In the method, a plurality of frames is first retrieved from the video. An image processing approach is performed to recognize image features of every frame. A plurality of color blocks for every frame are obtained through the image processing scheme. A variance among the pixels in every color block of two continuous frames can be obtained by a comparison computation, so as to form a color block variance. The color block variance allows the system to determine a motion target object as well as its image contour. The blocks excepting the blocks of the target object are set as transparent blocks. Therefore the background relative to the target object can be removed.

Abstract: A driver includes a display memory that stores display data, and a control unit that writes the display data in the display memory. The control unit accepts image region information for designating an image region in which an image corresponding to the display data is to be displayed within a display region of a display panel, and instruction information for giving an instruction to display of a frame region which surrounds the image region in the display region, and performs frame write processing for writing given tone data at an address of the display memory corresponding to the frame region, based on the image region information.

Abstract: A visual query is received from a client system, along with location information for the client system, and processed by a server system. The server system sends the visual query and the location information to a visual query search system, and receives from the visual query search system enhanced location information based on the visual query and the location information. The server system then sends a search query, including the enhanced location information, to a location-based search system. The search system receives and provides to the client one or more search results to the client system.

Abstract: A computer-implemented technique includes techniques are presented for user image capture feedback for improved machine language translation. When machine language translation of OCR text obtained from an initial image has a low degree of likelihood of being an appropriate translation, these techniques provide for user image capture feedback to obtain additional images to obtain a modified OCR text, which can result in improved machine language translation results. Instead of user image capture feedback, the techniques may obtain the modified OCR text by selecting another possible OCR text from the initial OCR operation. In addition to additional image capturing, light source intensity and/or a quantity/number of light source flashes can be adjusted. After obtaining the modified OCR text, another machine language translation can be obtained and, if it has a high enough degree of likelihood, it can then be output to a user.

Abstract: Systems, computer program products, and techniques for detecting objects depicted in digital image data are disclosed, according to various exemplary embodiments. The inventive concepts uniquely utilize internal features to accomplish object detection, thereby avoiding reliance on detecting object edges and/or transitions between the object and other portions of the digital image data, e.g. background textures or other objects. The inventive concepts thus provide an improvement over conventional object detection since objects may be detected even when edges are obscured or not depicted in the digital image data. In one aspect, a computer-implemented method of detecting an object depicted in a digital image includes: detecting a plurality of identifying features of the object, wherein the plurality of identifying features are located internally with respect to the object; and projecting a location of one or more edges of the object based at least in part on the plurality of identifying features.

Abstract: A method is disclosed including a print controller receiving print job data defined according to a first language, receiving a job ticket defined according to a second language, processing the print job data according to the second language and implementing counters defined according to the first language to track progress while processing the print job data.

Abstract: Differing embodiments of this disclosure may employ one or all of the several techniques described herein to perform credit card recognition using electronic devices with integrated cameras. According to some embodiments, the credit card recognition process may comprise: obtaining a first representation of a first image, wherein the first representation comprises a first plurality of pixels; identifying a first credit card region within the first representation; extracting a first plurality of sub-regions from within the identified first credit card region, wherein a first sub-region comprises a credit card number, wherein a second sub-region comprises an expiration date, and wherein a third sub-region comprises a card holder name; generating a predicted character sequence for the first, second, and third sub-regions; and validating the predicted character sequences for at least the first, second, and third sub-regions using various credit card-related heuristics, e.g.

Abstract: A server system receives a visual query and information about the current location of the querying device. The server system sends the query and location information to a visual query search system and receives one or more search results in accordance with the visual query and the location information. From the one or more search results, the server system identifies an entity in the visual query. The server system further identifies a client-side action corresponding to the identified entity and creates an actionable search result element configured to launch the client-side action. For example, if the identified entity were a restaurant the actionable search result elements could include buttons to initiate a telephone call, to read reviews, to map the address, or to make a reservation. If the actionable search result is selected, the client system launches the client-side action in an application distinct from the visual query client application.

Abstract: The production of a cutting pattern of a graphic placed upon a cutting mat is disclosed. A source image of the graphic overlaid on a cutting mat is received, and includes a plurality of registration marks as well as one or more distortions introduced during acquisition. The registration marks are identified from the source image by matching candidate sets of a plurality of center points of regions of adjacent groupings of pixels within the source image against predetermined positional relationships thereof corresponding to an actual arrangement of the registration marks on the cutting mat. An inverse transformation of the source image with values derived from the registration marks is applied. A corrected image aligned to physical coordinates of the cutting mat and referenced to the cutting machine is generated. A cut path is defined from vectors of the corrected image, and transmitted to the cutting machine for execution thereon.

Abstract: Musical compositions are generated from image pixels. To do so, pixel values are mapped to musical elements together for creating the musical compositions. Additionally, images are formed from pixels generated from musical compositions. More generally, a computer system creatively generates media using captured media as a source. The system also generates collage images in which individual images are pixels for the collage image. Collages are further generated from text.

Abstract: A computer determines whether performance optimization parameters are configured to optimize performance of generating a regular expression. In response to the computer determining the one or more performance optimization parameters are configured to optimize performance of generating the regular expression, the computer identifies syllables within a plain text word that have a high probability of spelling errors. The computer selects each character in the syllables identified. The computer identifies a group of characters from a confusion matrix that are commonly confused with the character selected. The computer generates a set of characters for each character selected, wherein the set of characters begin with one of the each character selected followed by and ending with the group of characters from the confusion matrix. The computer generates a regular expression by concatenating each of the set of characters.

Abstract: Embodiments of the subject technology provide for a hybrid OCR approach which combines server and device side processing that can offset disadvantages of performing OCR solely on the server side or the device side. More specifically, the subject technology utilizes image characteristics such as glyph details and image quality measurements to opportunistically schedule OCR processing on the mobile device and/or server. In this regard, text extracted by a “faster” OCR engine (e.g., one with less latency) is displayed to a user, which is then updated by the result of a more accurate OCR engine (e.g., an OCR engine provided by the server). This approach allows factoring in additional parameters such as network latency and user preference for making scheduling decisions. Thus, the subject technology may provide significant gains in terms of reduced latency and increased accuracy by implementing one or more techniques associated with this hybrid OCR approach.

Abstract: A computer receives a formatted query having a plain text word. The computer selects each character in the plain text word. The computer identifies a group of characters from a confusion matrix that are commonly confused with the character selected. The computer generates a set of characters for each character selected, wherein the set of characters begin with one of the each character selected followed by and ending with the group of characters from the confusion matrix. The computer generates a regular expression by concatenating each of the set of characters.

Abstract: The present invention concerns a method for building a watermarked content for sending to at least one user unit having a user unit identifier, the watermarked content comprising a first series of packets, at least some from the first series of packets being available in at least two different qualities, wherein said method comprises the steps of: generating a watermark based on at least the user unit identifier, the watermark being formed by a first series of values from a predetermined set of values; mapping each of the at least two different qualities to one from the predetermined set of values; building the watermarked content by sequentially selecting from the first series of packets, a packet having a quality corresponding to each value from the first series of values forming the watermark, It further concerns a method for transmitting such a watermarked content.

Abstract: An imaging device (100) is provided with: an imager (1) that includes an imaging element with multiple pixels arranged two-dimensionally, and that outputs an image signal in accordance with setting of either a pixel addition readout method or a cutout readout method; a signal processor (2) that processes the image signal from the imager (1); an operator (3) that inputs operation information; a lightness detector (4) that detects a subject lightness; and a controller (5). The controller (5) includes a zoom setter (51) that sets a zoom magnification for a subject based on a zoom signal that is input by the operator (3), and a readout-method switching controller (52) that switches between the pixel addition readout method and the cutout readout method according to the zoom magnification set by the zoom setter (51) and the lightness detected by the lightness detector (4).

Abstract: A method is provided that creates a lecture video capsule containing highlights of an original instructional video based on visual quality and content. The method includes segmenting and recognizing activities in the instructional video using a hidden Markov model (HMM). The activities are classified into three categories: talking head, writing hand and slideshow. The talking head frames are classified as non-content frames, while the writing hand and slideshows are classified as content frames. A non-reference based objective quality assessment of the non-content frames may be performed to detect high quality frames. Statistical parameters of an intensity histogram and a horizontal projection profile (HPP) of the content frames may be used to derive an objective quality measure of the content frames that is used to extract high quality content frames. The selected high quality non-content and content frames form a video clip or capsule, which is a temporally compressed representation of the video.