Top-Down and Bottom-Up Cues Based Moving Object Detection for Varied Background Video Sequences

Table 4

Comparison of accuracy of proposed approaches.

Video name	Total frames	Method 1	Method 2	Method 3	Method 4	Method 5	Method 6	Method 7	GMM [10]	KDE [15]
ATM	728	651	511	597	604	540	623	674	487	503
Cam	4567	4329	4018	4111	3986	3898	4178	4415	3795	3852
Campus_raw	1179	1094	993	1021	956	997	1024	1099	885	924
EPC-200-day-bird	1115	1005	994	973	989	898	1023	1097	774	855
HALLWAY_A	661	606	545	506	469	511	595	614	363	425
highwayI_raw	440	389	328	298	305	340	370	398	286	234
person15_walking_d1_uncomp	741	689	602	624	597	611	662	704	583	604
Stair	707	644	578	603	592	612	630	675	520	563
visor_1197284015696_07_fumogeno4	2024	1965	1901	1898	1830	1920	1946	1990	1719	1756
Total	12162	11435	10470	10631	10328	10327	11051	11739	9412	9716
Accuracy (%)		94.02	86.09	87.41	84.92	84.91	90.86	96.52	77.39	79.89