com Abstract Cascade detectors have been shown to operate extremely rapidly with high ac curacy and have important applications such as face detection Driven by this success cascade learning has been an area of active research in recent years Nev ert ID: 22741
Download Pdf The PPT/PDF document "MultipleInstance Pruning For Learning Ef..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
0 5 10 Positive Windows Negative Windows Cumulative Score Final Threshold 0 100 200 300 400 500 600 700 -20 -15 -10 -5 Feature Index Cumulative Score Final Threshold Positive windows but below threshold Positive windows above threshold Positive windows retained after pruning (a)(b) Final Threshold-3.0-2.5-2.0-1.5-1.0-0.50.0Detection Rate95.2%94.6%93.2%92.5%91.7%90.3%88.8% # of False Positive95513220875DBP36.1335.7835.7634.9329.2228.9126.72MIP16.1116.0616.8018.6016.9615.5314.59 ApproachViola-JonesBoosting chainFloatBoostWaldBoostWuet al.Soft cascadeTotal # of features606170025466007564943Slowness1018.118.913.9N/A37.1 (25)(a)(b) 0.9010.9030.9050.9070.909 Detection Rate MIP, T=-2.5, #f=11.25 B - B, alpha= - 16, #f=8.46 0.8950.8970.8991000110012001300140015001600 Detection Rate Number of False Positive B - B, alpha= - 16, #f=8.46 B-B, alpha=-8, #f=10.22 B-B, alpha=-4, #f=13.17 B-B, alpha=0, #f=22.75 B-B, alpha=4, #f=56.83