Human Computer Interaction Query by Sketch .


31 views
Uploaded on:
Description
Human Computer Interaction Query by Sketch. Chi-Ren Shyu. Department of Computer Engineering and Computer Science University of Missouri-Columbia Columbia, MO 65211, U.S.A. Query by Sketch -- an HCI topic?.
Transcripts
Slide 1

Human Computer Interaction Query by Sketch Chi-Ren Shyu Department of Computer Engineering and Computer Science University of Missouri Columbia, MO 65211, U.S.A.

Slide 2

Query by Sketch - a HCI theme? Human side : inquiry conduct, recovery result assessment, preparing of semantics and database ordering. PC side : question preparing from human\'s representations, include extraction from draw objects, database seek components, and so forth

Slide 3

Some Applications and Systems Photo (common scene pictures) IBM QBIC http://wwwqbic.almaden.ibm.com/GIS – Egenhofer at U. of Maine Spatial Query by portray http://www.spatial.maine.edu/~max/RL.html Medical Image – University of Missouri – Colmbia WebHIQS http://diglib1.cecs.missouri.edu:3243/FinalProj/index1.html (experimental site)

Slide 4

QBS in Diagnostic Image Databases Diagnoses regularly include discoveries identified with spatial relationship among injuries and historic points. Pathologies could be identified in light of numerous visual examples that are "sketchable, for example, states of tumors, dissemination of knobs, ect. Inquiry: "Recover database pictures that have comparative sore point of interest area with the outlined question (picture)"

Slide 5

Query techniques in Diagnostic Image DBS Sketch on a clear sheet – clients are given with a GUI to drawing organs, historic points, and sores. Portray on an organ layout – clients are furnished with a format of an organ and a GUI for drawing historic points and sores. Outline on a current picture – clients are furnished with a genuine indicative picture and a GUI for drawing historic points and injuries.

Slide 6

Research opportunity – Feature Extraction from Sketches Shapes from draw objects, surface from portrayed areas, dark scale from outlined region, and so on. Area skill will outline PC vision calculations for this reason. Framing a component vector for picture recovery. Hunting a high-dimensional database down recovering the most comparative representations?

Slide 7

Research opportunity – Spatial Modeling from Sketches Anatomical points of interest in HRCT lung: Lung areas (Automatic Extraction) Fissures (Human on the up and up) Lobe parcels (HIL/AE) Lesions versus points of interest: Interior Adjacent Across

Slide 8

Research opportunity – Spatial Modeling from Sketches - 2 Query injury Sketched by the client Why these three sores ought to be recovered?

Slide 9

Degree of Truth for (internal Adjacent, to one side) (0.34, 0.00) (0.15, 0.00) (0.55, 0.31) (1.00, 0.82) (1.00, 0.96) (0.83, 0.84) (0.21, 0.99) (0.00, 1.00) Spatial Modeling from Sketches - 3 What about various historic points with numerous sores?

Slide 10

Research opportunity – Semantic Query from Sketches Semantic or alleged philosophy is extremely subjective and client situated. Not useful for "general purposed" database recovery. Awesome for Sketches! The trust is: Sketches may have the capacity to bring more alluring recovery comes about because of client\'s angles.

Slide 11

Semantic Query from Sketches - 2

Slide 12

Semantic Query from Sketches - 3 Sketch bronchial structures on an organ format and fill dark scales to various part of the protest, for example, bronchial Walls and lumens.

Slide 13

Semantic Query from Sketches - 4 Users draw on numerous preparation pictures and frame a semantic tree structure for future recovery.

Slide 14

Semantic Query from Sketches - 5 For fluffy packs, this looks well known. We can shape enrollment capacities for the semantic terms by utilizing client\'s Sketches.

Slide 15

Semantic Query from Sketches - 6 Ranking database pictures by semantics

Slide 16

Semantic Query from Sketches - 6

Slide 17

Research opportunity – Query Languages for Sketches SELECT I.iid FROM LungImageDatabase I WHERE SIM( SKETCH(Qid,GUI), I.FE(iid)) < edge ORDER BY SIM; A conceivable SQL expansion for QBS:

Slide 18

Summary Query by portray is not new, be that as it may, it has high potential for some applications, particularly in the zones of indicative picture database, GIS, criminal examination, and so forth. In the CECS division, we have numerous current research extends that are exceedingly connected to the idea of QBS. We ought to have some kind of synergistic endeavors to make them intrigue extends sooner rather than later.

Recommended
View more...