If you use this dataset, please cite the following papers:
Yao G, Lei T, Zhong J, Jiang P, Jia W. Comparative Evaluation of Background Subtraction Algorithms in Remote Scene Videos Captured by MWIR Sensors. Sensors. 2017; 17(9):1945.
./RemoteSceneIRDataset: Remote Scene IR Dataset
IRVideoSequence: Frames caputered by MWIR camera
GroundTruth: Pixel-wise GroundTruth of foreground
./BSResults: Detected foreground masks by BS
./BSResultsM: Detected foreground masks by BS+MedianFilter
./BSResultsMM: Detected foreground masks by BS+MedianFilter+MorphologicalOperation
./BSResultsBGSLibrary Results of the BS from the BGSlibrary
16 BS algorithms are evaluated in Remote Scene IR Dataset:
AdaptiveMedian[1], Bayes[2], Codebook[3,4], Gaussian[5], GMG[6], GMM1[7], GMM2[8], GMM3[9], KDE[10], KNN[9], PBAS[11], PCAWS[12], Sigma-delta[13], SOBS[14], Texture[15], ViBe[16]
The implementations and parameter settings are detailed in the manuscript.
For a extensive evaluation, we also evaluate 24 BS algorithms implemented in BGSLibrary[17]
[1] N.J.B. McFarlane et al., Segmentation and tracking of piglets in images
[2] L. Li et al., Foreground object detection from videos containing complex background
[3] K. Kim et al., Background modeling and subtraction by codebook construction
[4] K. Kim et al., Real-time foreground–background segmentation using codebook model
[5] C.R. Wren et al., Pfinder: real-time tracking of the human body
[6] A.B. Godbehere et al., Visual tracking of human visitors under variable-lighting conditions for a responsive audio art installation
[7] C. Stauffer et al, Adaptive background mixture models for real-time tracking
[8] P. KaewTraKulPong et al., An improved adaptive background mixture model for real-time tracking with shadow detection
[9] Z. Zivkovic et al., Efficient adaptive density estimation per image pixel for the task of background subtraction
[10] A. Elgammal et al., Non-parametric model for background subtraction
[11] M. Hofmann et al., Background segmentation with feedback: the pixel-based adaptive segmenter
[12] P. St-Charles et al., Universal background subtraction using word consensus models
[13] A. Manzanera et al., A new motion detection algorithm based on Sigma-Delta background estimation
[14] L. Maddalena et al., A self-organizing approach to background subtraction for visual surveillance applications
[15] M. Heikkila et al., A texture-based method for modeling the background and detecting moving objects
[16] O. Barnich et al., ViBe: a universal background subtraction algorithm for video sequences
[17]A. Sobral, BGSLibrary: an opencv c++ background subtraction library