Method SCAD on the SBM-RGBD dataset

Contact name: Tsubasa Minematsu

Contact email: minematsu@limu.ait.kyushu-u.ac.jp

Contact University: Kyushu University

Method processing time: ~2 fps for a 640x480 video with C++ code running on 3.70 GHz Intel Xeon processor

Method parameters: k = 2.5, sigma_{a,f,ds,as}={160,2,1024,20}, ViBe_{N,R,min}={20,6,2}, SILTP_{N,R,tau} ={8,3,0.1} th_{g,H,S,V}={0.2,0.75,0.2,0.3}, alpha = 0.25)

Reference: T. Minematsu, A. Shimada, H. Uchiyama and R.-i. Taniguchi, Simple Combination of Appearance and Depth for Foreground Segmentation,in Battiato S., Farinella G., Leo M., Gallo G. (eds) New Trends in Image Analysis and Processing - ICIAP 2017. Lecture Notes in Computer Science, vol 10590, Springer, pp. 266--277, 2017.

Get all result files: here.

A) Sequence-by-sequence results

Name	Recall	Specificity	FPR	FNR	PWC	Precision	F-Measure
Bootstrapping
BootStrapping_ds	0.9796	0.9965	0.0035	0.0204	0.5120	0.9664	0.9729
adl24cam0	0.8604	0.9981	0.0019	0.1396	1.0240	0.9672	0.9107
bear_front	0.8326	0.9900	0.0100	0.1674	5.1480	0.9674	0.8949
fall01cam0	0.9312	0.9985	0.0015	0.0688	0.5259	0.9730	0.9516
fall20cam0	0.8950	0.9868	0.0132	0.1050	1.7975	0.7857	0.8368
ColorCamouflage
Cespatx_ds	0.9870	0.9993	0.0007	0.0130	0.0944	0.9674	0.9771
Hallway	0.9719	0.9979	0.0021	0.0281	0.3458	0.9622	0.9671
colorCam1	0.9935	0.9921	0.0079	0.0065	0.7604	0.9731	0.9832
colorCam2	0.9976	0.9725	0.0275	0.0024	1.6140	0.9678	0.9825
DepthCamouflage
DCamSeq1	0.9758	0.9958	0.0042	0.0242	0.5338	0.9318	0.9533
DCamSeq2	0.9777	0.9953	0.0047	0.0223	0.5468	0.9050	0.9399
Despatx_ds	0.9919	0.9956	0.0044	0.0081	0.4735	0.9575	0.9744
Wall	0.9910	0.9985	0.0015	0.0090	0.2189	0.9846	0.9878
IlluminationChanges
ChairBox	0.8801	0.9979	0.0021	0.1199	1.4947	0.9807	0.9277
Ls_ds	0.0000	1.0000	0.0000	0.0000	0.0000	0.0000	0.0000
TimeOfDay_ds	0.0000	0.9990	0.0010	0.0000	0.0957	0.0000	0.0000
genSeq1	0.9997	0.9738	0.0262	0.0003	2.2956	0.8461	0.9165
IntermittentMotion
Shelves	0.7823	0.9993	0.0007	0.2177	0.4965	0.9562	0.8605
Sleeping_ds	0.9878	0.9652	0.0348	0.0122	3.0269	0.8771	0.9292
abandoned1	0.9935	0.9978	0.0022	0.0065	0.2262	0.9222	0.9565
abandoned2	0.9948	0.9949	0.0051	0.0052	0.5079	0.9644	0.9794
movedBackground1	0.9865	0.9989	0.0011	0.0135	0.1301	0.9298	0.9573
movedBackground2	0.9928	0.9921	0.0079	0.0072	0.7818	0.8962	0.9420
OutOfRange
MultiPeople1	0.9696	0.9930	0.0070	0.0304	0.8850	0.9214	0.9449
MultiPeople2	0.9771	0.9962	0.0038	0.0229	0.5398	0.9576	0.9672
TopViewLab1	0.9527	0.9973	0.0027	0.0473	0.4206	0.9234	0.9378
TopViewLab2	0.9374	0.9974	0.0026	0.0626	0.4902	0.9353	0.9364
TopViewLab3	0.8062	0.9989	0.0011	0.1938	0.5202	0.9408	0.8683
Shadows
Shadows_ds	0.9797	0.9957	0.0043	0.0203	0.5293	0.9371	0.9579
fall01cam1	0.8567	0.9956	0.0044	0.1433	1.3178	0.9285	0.8912
genSeq2	0.9981	0.9880	0.0120	0.0019	1.0848	0.9128	0.9535
shadows1	0.9986	0.9880	0.0120	0.0014	1.0552	0.9312	0.9637
shadows2	0.9996	0.9878	0.0122	0.0004	1.0595	0.9285	0.9627

B) Average results for each category

Name	Recall	Specificity	FPR	FNR	PWC	Precision	F-Measure
Bootstrapping	0.8997	0.9940	0.0060	0.1003	1.8015	0.9319	0.9134
ColorCamouflage	0.9875	0.9904	0.0096	0.0125	0.7037	0.9677	0.9775
DepthCamouflage	0.9841	0.9963	0.0037	0.0159	0.4432	0.9447	0.9638
IlluminationChanges	0.4699	0.9927	0.0073	0.0301	0.9715	0.4567	0.4610
IntermittentMotion	0.9563	0.9914	0.0086	0.0437	0.8616	0.9243	0.9375
OutOfRange	0.9286	0.9965	0.0035	0.0714	0.5711	0.9357	0.9309
Shadows	0.9665	0.9910	0.0090	0.0335	1.0093	0.9276	0.9458
AVG.	0.8847	0.9932	0.0068	0.0439	0.9088	0.8698	0.8757