Abstract We present a method for foreground/background video segmentation (change detection) in real-time that can be used, in applications such as Background Subtraction or Analysis of Surveillance Cameras. Our approach implements a probabilistic segmentation based on the Quadratic Markov Measure Fields model. This framework regularizes the likelihood of each pixel belonging to each one of the classes (background or foreground). We propose a new likelihood that takes into account two cases: the first one is when the background is static and the foreground might be static or moving [Static Background Subtraction (SBS)], the second one is when the background is unstable and the foreground is moving [Unstable Background Subtraction (UBS)]. Moreover, our likelihood is robust to illumination changes, cast shadows and camouflage situations. We implement a parallel version of our algorithm in CUDA using a NVIDIA GPU in order to fulfill real-time execution requirement. |
No. |
Test Seq. |
Original Video |
Proposal using Graph Cut | Proposal using QMMF |
1 |
Moved
Object |
MO_Original.avi |
MO_Segmented_GraphCut.avi |
MO_Segmented_QMMF.avi |
2 |
Time Of
Day |
TOD_Original.avi |
TOD_Segmented_GraphCut.avi |
TOD_Segmented_QMMF.avi |
3 |
Light
Switch |
LS_Original.avi |
LS_Segmented_GraphCut.avi |
LS_Segmented_QMMF.avi |
4 |
Waving
Trees |
WT_Original.avi |
WT_Segmented_GraphCut.avi |
WT_Segmented_QMMF.avi |
5 |
Camouflage |
C_Original.avi |
C_Segmented_GraphCut.avi |
C_Segmented_QMMF.avi |
6 |
Bootstrapping |
B_Original.avi |
B_Segmented_GraphCut.avi |
B_Segmented_QMMF.avi |
7 |
Foreground
Aperture |
FA_Original.avi |
FA_Segmented_GraphCut.avi |
FA_Segmented_QMMF.avi |
No. |
Test Seq. |
Original Video |
Proposal using Graph Cut | Proposal using QMMF |
1 |
Video2_long |
V2_Original.avi |
V2_Segmented_GraphCut.avi |
V2_Segmented_QMMF.avi |
2 |
Video3_long | V3_Original.avi |
V3_Segmented_GraphCut.avi |
V3_Segmented_QMMF.avi |
3 |
Video4_long | V4_Original.avi |
V4_Segmented_GraphCut.avi |
V4_Segmented_QMMF.avi |
4 |
Video5_long | V5_Original.avi |
V5_Segmented_GraphCut.avi |
V5_Segmented_QMMF.avi |
5 |
Video6_long | V6_Original.avi |
V6_Segmented_GraphCut.avi |
V6_Segmented_QMMF.avi |
6 |
Video7_long | V7_Original.avi |
V7_Segmented_GraphCut.avi |
V7_Segmented_QMMF.avi |
7 |
Video8_vlong | V8_Original.avi |
V8_Segmented_GraphCut.avi |
V8_Segmented_QMMF.avi |