1 / 72

Multimedia/Video Search

Multimedia/Video Search. חיפוש מולטימדיה בדגש חוזי. אריאל פרנק מחלקה למדעי המחשב אוניברסיטת בר-אילן ariel@cs.biu.ac.il. Contents. Multimedia (MM) and search/retrieval Text-based MM search in General SEs Text-based MM search in Vertical SEs Tag-based MM search Content-based MM search

hetal
Download Presentation

Multimedia/Video Search

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Multimedia/Video Search חיפוש מולטימדיה בדגש חוזי אריאל פרנק מחלקה למדעי המחשב אוניברסיטת בר-אילן ariel@cs.biu.ac.il A. Frank

  2. Contents • Multimedia (MM) and search/retrieval • Text-based MM search in General SEs • Text-based MM search in Vertical SEs • Tag-based MM search • Content-based MM search • Video downloads A. Frank

  3. Multimedia Resources • תמונות (Pictures, Photos, Maps, Graphics,Animations) • שמע (Audio, Sounds) • מוסיקה(Music, MP3) • הקלטות (Radio, Podcasts, Speeches) • חוזי(Video, Flash, Vlogs) • סרטים (Movies, TV) A. Frank

  4. MM Search & Information Retrieval (MMIR) • Search & information retrieval of multimedia objects is hard!! • Using SEs to find relevant, quality MM objects can be difficult and frustrating. • Sample query: find a movie that has a large mouse chasing a Persian cat with twins watching from above? A. Frank

  5. MM Search & Information Retrieval (MMIR) • Common MM search is mainly text-based; the object text is derived from: • caption of the MM object • text surrounding the MM object • entire text of the containing page • filenames of MM object and the containing page • manual/automatic annotations • Recently, tag-based MM search is becoming effective (social computing). • Fortunately, long due, content-based MM search on the Web is forthcoming. A. Frank

  6. Contents • Multimedia (MM) and search/retrieval • Text-based MM search in General SEs • Text-based MM search in Vertical SEs • Tag-based MM search • Content-based MM search • Video downloads A. Frank

  7. General SEs Sample • Google (general, video) • Yahoo (general, audio) • AOL (general, video) • MSN/Live (images, video) • AltaVista (images, audio, video) A. Frank

  8. Multimedia: Google A. Frank

  9. Multimedia: Google A. Frank

  10. Multimedia: Yahoo A. Frank

  11. Yahoo Audio Search A. Frank

  12. A. Frank

  13. AOL Video Search A. Frank

  14. MSN Live A. Frank

  15. Multimedia: MSN Live A. Frank

  16. AltaVista Multimedia Search A. Frank

  17. Contents • Multimedia (MM) and search/retrieval • Text-based MM search in General SEs • Text-based MM search in Vertical SEs • Tag-based MM search • Content-based MM search • Video downloads A. Frank

  18. Vertical SEs Sample • Picsearch (image) • FindSounds (audio) • SearchVideo (video) • Clipblast! (video) • Flurl (multimedia) A. Frank

  19. Images: Picsearch A. Frank

  20. Picsearch Advanced Interface A. Frank

  21. Audio: FindSounds A. Frank

  22. Types of Sounds A. Frank

  23. Video: SearchVideo A. Frank

  24. Video: Clipblast! • Organizes and makes the video Web relevant, fast and simple to navigate. • Video Search and Navigation technology makes it easy to search, browse and personalize the video that viewers want, when they want it. • Crawls the entire Web for all video content available, indexing more video content providers than any other video search engine. • Largest video distribution platform ever. A. Frank

  25. Video: Clipblast A. Frank

  26. Multimedia: Flurl A. Frank

  27. Flurl list of sites A. Frank

  28. Contents • Multimedia (MM) and search/retrieval • Text-based MM search in General SEs • Text-based MM search in Vertical SEs • Tag-based MM search • Content-based MM search • Video downloads A. Frank

  29. Tag-based Search Sample • Flickr (image) • Panoramio (image, maps) • YouTube (video) • Keotag (multimedia) A. Frank

  30. Flickr A. Frank

  31. Flickr A. Frank

  32. Flickr A. Frank

  33. Image/Maps: Panoramio A. Frank

  34. Image/Maps: Panoramio A. Frank

  35. Image/Maps: Panoramio A. Frank

  36. Video: YouTube A. Frank

  37. Keotag Meta-SE A. Frank

  38. Keotag Meta-SE A. Frank

  39. Contents • Multimedia (MM) and search/retrieval • Text-based MM search in General SEs • Text-based MM search in Vertical SEs • Tag-based MM search • Content-based MM search • Video downloads A. Frank

  40. “Content-based“ Image Search/Retrieval • “Content based” image DB or Web site(s)that contain a large number of images. • Image analysis techniques used to describe images by their extracted visual features such as color, texture, shape and orientation. • These objects can be represented by sets of image features and then compared with the sets extracted from other images. • Feature sets are described by their statistics, and the computer searches for statistically likely matches. • Queries can be formulated in terms of visual features or by similarity options. A. Frank

  41. Content-based Search Sample • Squid (image) • retriever (image) • Melodyhound (audio) • Musipedia (music) • Blinkx (video) A. Frank

  42. Image: Squid • Has about 1100 images of marine creatures in its database. • Each image shows one distinct species on a uniform background. • Every image is processed to recover the boundary contour, which is then represented by three global shape parameters and the maxima of the curvature zero-crossing contours in its Curvature Scale Space (CSS) image. A. Frank

  43. Images: Squid A. Frank

  44. Images: Squid A. Frank

  45. Images: Squid A. Frank

  46. Images: Squid A. Frank

  47. Image: retriever • Experimental service which lets you search and explore in a selection of Flickr images by drawing a rough sketch. • Matches the most pronounced shapes and slabs of colors. • The results are usually fairly good, sometimes even stunning, • But it doesn't do object/face recognition of any kind, so if drawing an outline sketch of a chair, it almost certainly won't get you one back. A. Frank

  48. Images: retriever A. Frank

  49. Images: retriever A. Frank

  50. Images: retriever A. Frank

More Related