TUB-IRML at the MediaEval 2014 Visual Privacy Task

Data & Analytics

esra-acar
of 17
Description
Text
  1. 1. TUB-IRML at MediaEval 2014 Visual Privacy Task:Privacy Filtering through Blurring and Color RemappingDominique Maniry, Esra Acar, Sahin AlbayrakCompetence Center Information Retrieval & Machine Learning
  2. 2. Outline►The Privacy Filter►Sample Outputs of the Filter►Discussion on the Filter►Performance Evaluation►Conclusions & Future Work17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 2
  3. 3. The Privacy Filter (1)►Main idea: To obscure both shape and appearance ofidentity-related regions through blurring and colorremapping.►Preserve the intelligibility bydisplaying edges, andhinting anomalous events through special colors.17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 3
  4. 4. The Privacy Filter (2)►The filter contains four steps:Step 1: Blur all privacy-related regionsStep 2: Reduce number of colors & remap colorsStep 3: Apply a blending maskStep 4: Include shape information by incorporatingedges17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 4
  5. 5. Step 1: Blur17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 5
  6. 6. Step 2: Reduce Colors17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 6
  7. 7. Step 2: Remap Colors17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 7
  8. 8. Step 3: Apply a Blending Mask►The blending mask mask(x, y) is a binary image whereannotated regions have a value of 1 and remainingregions have a value 0.►The smoothing is achieved by applying a Gaussian blurto the blending mask.17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 8
  9. 9. Step 4: Include Shape Information (1)►The obscured regions are overlaid with edgesobtained with Canny Edge detection.►Edges in regions with a high privacy requirement (i.e.,faces) are discarded.►The remaining edges are emphasized usingmorphological dilation with a 3x3 circle as structuringelement.17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 9
  10. 10. Step 4: Include Shape Information (2)A walking person Two people fighting17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 10
  11. 11. Sample Outputs of the Filter (1)17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 11
  12. 12. Sample Outputs of the Filter (2)17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 12
  13. 13. Discussion on the FilterPros Cons Parameters to tune trade-offbetween privacy and intelligibility(blur intensity and number ofcolors). Remapped colors can conveyadditional information. Different regions can havedifferent privacy levels by usingdifferent blur intensities (e.g.,face more blurred than fullbody).17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 13 Simple. Identity related details can leakthrough shape.
  14. 14. Performance Evaluation (1)Stream 1: 230 crowd-sourcing workers.Stream 2: 65 people working at Thales (mainly in R&D).Stream 3: 59 participants from sectors including R&D, data protection andlaw enforcement from all around the world.17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 14
  15. 15. Performance Evaluation (2)Stream 1 Results Stream 2 ResultsStream 3 Results17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 15
  16. 16. Conclusions & Future Work►The user study has shown that our method is veryeffective at protecting privacy.►Future workEvaluating different parameters to balance privacy andintelligibility, and Improving the appropriateness by reducing the obscuredregions using a pixel-wise segmentation.17 October 2014 TUB-IRML at MediaEval 2014 Visual Privacy Task 16
  17. 17. M.Sc.Competence Center Information Retrieval &Machine Learningwww.dai-labor.deFonFax+49 (0) 30 / 314 – 74+49 (0) 30 / 314 – 74 003DAI-LaborTechnische Universität BerlinFakultät IV – Elektrontechnik & InformatikSekretariat TEL 14Ernst-Reuter-Platz 710587 Berlin, Deutschland17Esra AcarResearcheresra.acar@tu-berlin.deThanks!013TUB-IRML at 17 October 2014 MediaEval 2014 Visual Privacy Task
Comments
Top