(*) In the results reported in the 2012 MIT tech report and on the website prior to Sept. 25, 2014, histogram matching was performed on the saliency maps (to match them to the ground truth human maps) prior to computation of the SIM and EMD metrics. The numbers presented in this table are a result of histogram matching, and thus may overestimate the performance of these models. We have since removed histogram matching from evaluation because we consider it important to report the raw model performances, to make them replicable and reproducible (particulary when no human ground truth maps are available).
Model Name | Published | Code | AUC-1 [?] | SIM [?] | EMD [?] | AUC-2 [?] | sAUC [?] | CC [?] | NSS [?] | Date tested [key] |
---|---|---|---|---|---|---|---|---|---|---|
Baseline: infinite humans [?] | 0.9121 | 1 | 0 | 0.8682 | 0.8001 | 1 | 3.1810 | |||
Boolean Map based Saliency (BMS) | Jianming Zhang, Stan Sclaroff. Saliency detection: a boolean map approach [ICCV 2013] | matlab, executable | 0.8257 | 0.5364 | 3.2620 | 0.8152 | 0.6522 | 0.5499 | 1.4138 | first tested: 14/05/2014 last tested: 16/06/2014 maps from authors |
Mixture of Saliency Models | Xuehua Han, Shunji Satoh. "Unifying computational models for visual attention" [AINI 2014, Sep. (accepted)] | 0.8209 | 0.5232 | 3.0180 | 0.8122 | 0.6206 | 0.5194 | 1.3397 | first tested: 08/08/2014 last tested: 08/08/2014 maps from authors |
|
Ensembles of Deep Networks (eDN) | Eleonora Vig, Michael Dorr, David Cox. Large-Scale Optimization of Hierarchical Features for Saliency Prediction in Natural Images [CVPR 2014] | python | 0.8192 | 0.5123 | 3.0129 | 0.8111 | 0.6152 | 0.4549 | 1.1420 | first tested: 16/08/2014 last tested: 16/06/2014 maps from authors |
Outlier Saliency (OS) | Chuanbo Chen, He Tang, Zehua Lyu, Hu Liang, Jun Shang, Mudar Serem. Saliency Modeling via Outlier Detection [Journal of Electronic Imaging]. Accepted, 2014. | 0.8183 | 0.5195 | 3.2742 | 0.8079 | 0.6219 | 0.5364 | 1.3793 | first tested: 16/09/2014 last tested: 16/09/2014 maps from authors |
|
Judd Model | Tilke Judd, Krista Ehinger, Fredo Durand, Antonio Torralba. Learning to predict where humans look [ICCV 2009] | matlab | 0.8093 | 0.5018 | 3.1583 | 0.8011 | 0.5956 | 0.4665 | 1.1781 | last tested: 16/06/2014 maps from code (DL:17/12/2013) with default params |
CovSal | Erkut Erdem, Aykut Erdem. Visual saliency estimation by nonlinearly integrating features using region covariances [JoV 2013] | matlab | 0.8053 | 0.5018 | 3.1092 | 0.6693 | 0.5735 | 0.4518 | 1.2207 | first tested: 05/02/2012 last tested: 29/05/2014 maps from authors |
Fast and Efficient Saliency (FES) | Hamed Rezazadegan Tavakoli, Esa Rahtu, Janne Heikkila. Fast and efficient saliency detection using sparse sampling and kernel density estimation [SCIA 2011] | matlab | 0.8033 | 0.4952 | 3.3488 | 0.7343 | 0.5869 | 0.4846 | 1.2738 | first tested: 10/04/2013 last tested: 28/05/2014 maps from authors |
Graph-Based Visual Saliency (GBVS) | Jonathan Harel, Christof Koch, Pietro Perona. Graph-Based Visual Saliency [NIPS 2006] | matlab | 0.8066 | 0.4813 | 3.4359 | 0.7966 | 0.6290 | 0.4809 | 1.2413 | last tested: 27/06/2014 maps from code (DL:20/08/2013) with default params |
Baseline: one human [?] | 0.8011 min: 0.7609 max: 0.8339 |
0.4879 min: 0.3992 max: 0.5751 |
3.8146 min: 3.5021 max: 4.1368 |
0.6646 min: 0.6343 max: 0.7066 |
0.6350 min: 0.6037 max: 0.6715 |
0.5250 min: 0.3763 max: 0.6524 |
1.6455 min: 1.2056 max: 2.1001 |
|||
Sampled Template Collation | Andreas Holzbach, Gordon Cheng. A Scalable and Efficient Method for Salient Region Detection using Sampled Template Collation [ICIP 2014] | 0.7935 | 0.4757 | 3.4854 | 0.7836 | 0.5440 | 0.3981 | 0.9735 | first tested: 04/12/2013 last tested: 11/08/2014 maps from authors |
|
Region Contrast (RC) | Ming-Ming Cheng, Niloy J. Mitra, Xiaolei Huang, Philip H. S. Torr, Shi-Min Hu. Global Contrast based Salient Region detection [IEEE TPAMI 2014] Ming-Ming Cheng, Niloy J. Mitra, Xiaolei Huang, Philip H. S. Torr, Shi-Min Hu. Salient Object Detection and Segmentation [CVPR 2011] |
c++, executable | 0.7922 | 0.4705 | 3.4180 | 0.7800 | 0.5489 | 0.4709 | 1.1760 | first tested: 08/03/2013 last tested: 28/05/2014 maps from authors |
Multi-Resolution AIM (MR-AIM) | Siddharth Advani, John Sustersic, Kevin Irick, Vijaykrishnan Narayanan. A multi-resolution saliency framework to drive foveation [ICASSP 2013] | 0.7718 | 0.4711 | 3.3635 | 0.7632 | 0.5458 | 0.3892 | 0.9569 | first tested: 27/05/2013 last tested: 28/05/2014 maps from authors |
|
CWS model | Unpublished work | 0.7896 | 0.4638 | 3.7448 | 0.7846 | 0.5512 | 0.4529 | 1.1118 | first tested: 14/05/2014 last tested: 16/06/2014 maps from authors |
|
MKL-based model | Yasin Kavak, Aykut Erdem, Erkut Erdem. Visual saliency estimation by integrating features using multiple kernel learning [ISACS 2013] | 0.7836 | 0.4465 | 3.3212 | 0.7770 | 0.6114 | 0.4246 | 1.0797 | first tested: 17/03/2014 last tested: 28/05/2014 maps from authors |
|
Baseline: Center [?] | matlab | 0.7831 | 0.4491 | 3.7225 | 0.7723 | 0.5064 | 0.3809 | 0.9190 | ||
Saliency for Image Manipulation | Ran Margolin, Lihi Zelnik-Manor, Ayellet Tal. Saliency for Image Manipulation [CGI 2012] | matlab | 0.7741 | 0.4391 | 4.1368 | 0.7582 | 0.6397 | 0.4326 | 1.1390 | first tested: 01/07/2012 last tested: 28/05/2014 maps from authors |
RARE2012 | Nicolas Riche, Matei Mancas, Matthieu Duvinage, Makiese Mibulumukini, Bernard Gosselin, Thierry Dutoit. RARE2012: A multi-scale rarity-based saliency detection with its comparative statistical analysis [Signal Processing: Image Communication, 2013] | matlab | 0.7719 | 0.4363 | 4.1019 | 0.7542 | 0.6654 | 0.4245 | 1.1511 | first tested: 31/08/2012 last tested: 28/05/2014 maps from authors |
LMF | Liang Jiayu Gary. Unpublished work. | 0.7703 | 0.4360 | 4.8645 | 0.7619 | 0.6373 | 0.4137 | 1.0678 | first tested: 06/10/2013 last tested: 28/05/2014 maps from authors |
|
AIM | Neil Bruce, John Tsotsos. Attention based on information maximization [JoV 2007] | matlab | 0.7654 | 0.4176 | 4.1622 | 0.7536 | 0.6564 | 0.3120 | 0.7867 | last tested: 05/07/2014 maps from code (DL:15/01/2014) with params: resize=0.5, convolve=1, thebasis='31infomax975' |
Random Center Surround Saliency | Tadmeri Narayan Vikrama, Marko Tscherepanowa, Britta Wredea. A saliency map based on sampling an image into random rectangular regions of interest [Pattern Recognition 2012] | matlab | 0.7531 | 0.4200 | 3.4645 | 0.7366 | 0.5475 | 0.3778 | 0.9451 | first tested: 15/05/2012 last tested: 28/05/2014 maps from authors |
Image Signature | Xiaodi Hou, Jonathan Harel, Christof Koch. Image Signature: Highlighting Sparse Salient Regions [PAMI 2011] | matlab | 0.7528 | 0.4131 | 4.8002 | 0.7409 | 0.6560 | 0.3829 | 1.0094 | first tested: 19/06/2014 last tested: 23/06/2014 maps from authors |
IttiKoch2 | Implementation by Jonathan Harel (part of GBVS toolbox) | matlab | 0.7511 | 0.4012 | 4.2587 | 0.7423 | 0.6307 | 0.3742 | 0.9667 | last tested: 27/06/2014 maps from code (DL:20/08/2013) with default params |
Visual Conspicuity (VICO) | Matthieu Perreira Da Silva, Vincent Courboulay. Implementation and Evaluation of a Computational Model of Attention for Computer Vision [book chapter, 2012] | binaries | 0.7495 | 0.4147 | 4.7735 | 0.7069 | 0.6027 | 0.3713 | 0.9749 | first tested: 28/11/2012 last tested: 28/05/2014 maps from authors |
Context-Aware saliency | Stas Goferman, Lihi Zelnik-Manor, Ayellet Tal. Context-Aware Saliency Detection [CVPR 2010] [PAMI 2012] | matlab | 0.7415 | 0.3899 | 4.9000 | 0.7313 | 0.6548 | 0.3555 | 0.9472 | last tested: 24/06/2014 maps from code (DL:15/01/2014) with default params |
Adaptive Whitening Saliency Model (AWS) | Anton Garcia-Diaz, Victor Leboran, Xose R. Fdez-Vidal, Xose M. Pardo. On the relationship between optical variability, visual saliency, and eye fixations: A computational approach [JoV 2012] | matlab | 0.7397 | 0.4045 | 4.9308 | 0.7319 | 0.6790 | 0.3672 | 1.0139 | last tested: 24/06/2014 maps from code (DL:17/01/2014) with params: rescale=0.5 |
Weighted Maximum Phase Alignment Model (WMAP) | Fernando Lopez-Garcia, Xose Ramon Fdez-Vidal, Xose Manuel Pardo, Raquel Dosil. Scene Recognition through Visual Attention and Image Features: A Comparison between SIFT and SURF Approaches | matlab | 0.7357 | 0.3965 | 4.5788 | 0.6651 | 0.6273 | 0.3378 | 0.9684 | last tested: 27/06/2014 maps from code (DL:17/01/2014) with params: rescale=0.5 |
NARFI saliency | Jiazhong Chen, Hua Cao, Zengwei Ju, Leihua Qin, Shuguang Su. Non-attention region first initialisation of k-means clustering for saliency detection [Electronics Letters 2013] | 0.7258 | 0.4202 | 4.6974 | 0.6125 | 0.5499 | 0.3329 | 0.8307 | first tested: 05/11/2013 last tested: 28/05/2014 maps from authors |
|
Self-resemblance by LARK | Hae Jong Seo, Peyman Milanfar. Static and Space-time Visual Saliency Detection by Self-Resemblance [JoV 2012] | matlab | 0.7110 | 0.3608 | 4.9504 | 0.6941 | 0.6416 | 0.3093 | 0.8304 | first tested: 20/06/2014 last tested: 27/06/2014 maps from authors |
Murray model (Chromatic Induction Wavelet Model) | Naila Murray, Maria Vanrell, Xavier Otazu, C. Alejandro Parraga. Saliency Estimation Using a Non-Parametric Low-Level Vision Model [CVPR 2011] | matlab | 0.6962 | 0.3447 | 4.7261 | 0.6913 | 0.6521 | 0.2696 | 0.7298 | last tested: 27/06/2014 maps from code (DL:29/05/2014) with default params |
Quantum-Cuts (QCUT) | Caglar Aytekin, Serkan Kiranyaz, Moncef Gabbouj | 0.6901 | 0.3978 | 4.9924 | 0.6366 | 0.5500 | 0.2720 | 0.7111 | first tested: 19/12/2013 last tested: 28/05/2014 maps from authors |
|
Torralba saliency | Antonio Torralba, Aude Oliva, Monica S. Castelhano, John M. Henderson. Contextual Guidance of Attention in Natural scenes: The role of Global features on object search [Psychological Review 2006] | matlab | 0.6837 | 0.3438 | 4.7277 | 0.6769 | 0.6241 | 0.2524 | 0.6850 | last tested: 27/06/2014 maps from code (here) with default params |
Baseline: Permutation Control [?] | 0.6822 | 0.3402 | 4.5884 | 0.5853 | 0.5006 | 0.2025 | 0.4916 | |||
SUN saliency | Lingyun Zhang, Matthew H. Tong, Tim K. Marks, Honghao Shan, Garrison W. Cottrell. SUN: A Bayesian framework for saliency using natural statistics [JoV 2008] | matlab | 0.6719 | 0.3413 | 5.0753 | 0.6640 | 0.6137 | 0.2478 | 0.6832 | last tested: 27/06/2014 maps from code (DL:15/01/2014) with params: scale=0.5 |
IttiKoch | Implemented in the Saliency Toolbox by: Dirk Walther, Christof Koch. Modeling attention to salient proto-objects [Neural Networks 2006] | matlab | 0.5951 | 0.3056 | 4.7857 | 0.5405 | 0.5317 | 0.1435 | 0.4278 | last tested: 27/06/2014 maps from code (DL:15/01/2014) with params: sampleFactor='dyadic' |
Achanta | Radhakrishna Achanta, Sheila Hemami, Francisco Estrada, Sabine Susstrunk. Frequency-tuned Salient Region Detection [CVPR 2009] | matlab, c++, executable | 0.5220 | 0.2970 | 6.8617 | 0.5201 | 0.5217 | 0.0423 | 0.1310 | last tested: 24/06/2014 maps from code (DL:15/01/2014) with params: GausParam=[3,3] |
Baseline: Chance [?] | matlab | 0.5017 | 0.3277 | 6.3460 | 0.5000 | 0.4998 | 0.0001 | 0.0000 |