Image-based retrieval for high quality visual or acoustic rendering
The embodiments of the invention relate to method for image-based retrieval of an image or text, comprising capturing a digital image; matching the digital image against digitally-stored documents; and retrieving one or more of the digitally-stored documents that match the digital image. Other embodiments relate to a device for image-based retrieval and rendering, comprising an image capturing device for capturing digital image; a device for processing the digital image to match it against digitally-stored documents; a device for retrieving a digitally-stored document that matches the digital image; and a device for rendering the selected retrieved image.
1. A method for image-based retrieval of an image or text, comprising:
capturing at least a portion of a cover page and a digital image, the digital image comprising text, and the cover page identifying a publication source of the digital image;
matching, without optical character recognition, the digital image against digitally-stored documents;
retrieving one or more of the digitally-stored documents that match the digital image, the digitally stored documents comprising a composite of images in a defined layout; and
retrieving from the composite of images an article corresponding to the captured digital image using a greatest overlapping area between a geometry of the defined layout and a boundary of the digital image.
2. The method of claim 1 , wherein the image documents are retrieved using near-duplicate image detection, wherein near-duplicate image detection comprises indexing a plurality of distinctive local descriptors.
3. The method of claim 1 , further comprising rendering the retrieved image.
4. The method of claim 3 , wherein rendering includes converting the retrieved image for visual display.
5. The method of claim 3 , wherein rendering includes converting the retrieved image for audio output.
6. The method of claim 5 , wherein converting the retrieved image for audio output includes converting the image to text and converting the text to speech.
7. The method of claim 3 , wherein rendering includes converting the retrieved image for Braille output.
8. The method of claim 7 , wherein converting the retrieved image for Braille output includes converting the image to text.
9. The method of claim 1 , wherein the digital image is captured using a digital camera or a scanner.
10. A method for image-based retrieval and rendering, the method comprising:
capturing at least a portion of a cover page and a digital image, the digital image comprising text, and the cover page identifying a publication source of the digital image;
matching, without optical character recognition, the digital image against digitally-stored documents;
retrieving a digitally-stored document that matches the digital image, the digitally stored document comprising a composite of images in a defined layout;
retrieving from the composite of images an article corresponding to the captured digital image using a greatest overlapping area between a geometry of the defined layout and a boundary of the digital image; and
rendering the retrieved document.
11. The method of claim 10 , wherein rendering includes converting the retrieved image for visual display.
12. The method of claim 10 , wherein rendering includes converting the retrieved image for audio output.
13. The method of claim 12 , wherein converting the retrieved image for audio output includes converting the image to text and converting the text to speech.
14. The method of claim 10 , wherein rendering includes converting the retrieved image for Braille output.
15. The method of claim 14 , wherein converting the retrieved image for Braille output includes converting the image to text.
16. The method of claim 10 , wherein the image documents are retrieved using near-duplicate image detection.
17. The method of claim 10 , wherein the digital image is captured using a digital camera or a scanner.
18. A system for image-based retrieval and rendering, comprising:
an image capturing device configured to capture at least a portion of a cover page and a digital image comprising text, the cover page identifying a publication source of the text of the digital image;
a device configured to send the digital image to a search engine for retrieving one or more image documents that match the digital image;
a device configured to process the digital image to match the digital image, without optical character recognition, against digitally-stored documents, the digitally stored documents comprising a composite of images in a defined layout; and
a device configured to retrieve from the composite of images an article corresponding to the captured digital image using a greatest overlapping area between a geometry of the defined layout and a boundary of the digital image.
19. The system of claim 18 , wherein the image capturing device includes a digital camera or a scanner.
20. The system of claim 18 , wherein one or more computers process the digital image to match it against digitally-stored documents, retrieve the digitally stored document that matches the digital image, and render the selected retrieved image.
21. The system of claim 18 , wherein the image capture device is a digital camera or a scanner.
22. The system of claim 21 , further comprising a device for rendering the one or more images that match the digital image.
23. The method of claim 2 , wherein near-duplicate image detection further comprises using Lowe's Difference of Gaussian detector.
24. The method of claim 23 , further comprising constructing a Gaussian pyramid and searching for keypoints.