people

Ali Borji, University of Central Florida
Jack Brown, Massachusetts Institute of Technology
Zoya Bylinskii, Massachusetts Institute of Technology
Fredo Durand, Massachusetts Institute of Technology
Laurent Itti, University of Southern California
Tilke Judd, Google Zurich
Aude Oliva, Massachusetts Institute of Technology
Antonio Torralba, Massachusetts Institute of Technology

Updates

13/09/2016
- added LDS, I-VGG16 (UID), UHF, iSEEL to MIT300 results

1/09/2016
- added ML-Net to MIT300 results

23/06/2016
- added GMR and DSCLRCN models to MIT300 results

12/15/2016
- added visualizations for MIT300
- added visualizations for CAT2000

20/04/2016
- added paper to KalSal model

16/04/2016
- added code for the visualization of metrics

13/04/2016
- added new paper about MIT Saliency Benchmark: What do different evaluation metrics tell us about saliency models?

01/03/2016
- added JuntingNet, PDP, 2 anonymous CVPR 2016 models to MIT300 results

22/02/2016
- added Deep Gaze 2 to MIT300 results

03/02/2016
- added WEPSAM model to MIT300 results
- added SDDPM to CAT2000 results

27/01/2016
- added infinite observers baseline to CAT2000 results

26/01/2016
- added KalSal model to MIT300 results

25/01/2016
- added KL metric to CAT2000 results, for a total of 8 evaluation metrics
- added AIM model to CAT2000 results
- added per-category mean scores and standard deviations for each model on the CAT2000 results page (as separate downloadable CSV files, provided for additional analysis)

17/01/2016
- added FIGRIM fixation dataset to the database page

11/01/2016
- MR-AIM code now available and MIT300 Results updated

03/12/2015
- added KL metric to MIT300 results
- added Fabrice Urban, Mr-CNN, CNN-VLM to MIT300 results

03/11/2015
- pushed some metric code fixes to the GitHub repo
- fixed 3 images (one in trainSet, two in testSet) that were of different sizes in the CAT2000 database, images updated

16/10/2015
- DeepFix, eDN, RARE2012 and BMS added to CAT2000 results

16/10/2015
- DeepFix added to MIT300 results

11/08/2015
- Mr-CNN added to MIT300 results

28/07/2015
- Added Coutrot Database 1 and 2 to the database page

19/06/2015
- Rosin saliency and SalNet saliency added to MIT300 results

17/06/2015
- "CAT2000: A Large Scale Fixation Dataset for Boosting Saliency Research" will be presented at the CVPR 2015 workshop on "Future of Datasets"

02/05/2015
- renamed AUC-1 to AUC-Judd and AUC-2 to AUC-Borji for clarity
- added Aboudib Magnification Saliency to MIT300 and CAT2000

22/04/2015
- added additional baselines to CAT2000 results

07/04/2015
- updated results of QCUT saliency

20/03/2015
- added Co-Occurrence Histogram based Saliency
- added SALICON saliency

09/03/2015
- added GNM saliency
- added fixation locations (for possible scanpath analysis) to CAT2000 training images

11/02/2014
- Ali Borji and Laurent Itti have joined the Saliency Benchmark Team!
- CAT2000 dataset has been added to the saliency benchmark

27/12/2014
- added Stochastic fixation prediction (SFP) model

04/12/2014
- updated LGS and LS models to smoothed versions

27/11/2014
- added LGS and LS models

18/11/2014
- added SAVAM video dataset to other datasets

05/11/2014
- added arxiv links for Deep Gaze 1

22/10/2014
- added Deep Gaze 1 performance

03/10/2014
- added Aboudib Magnification Saliency

29/09/2014
- added SWT saliency
- added baseline sample images

25/09/2014
- changed evaluation protocol: omit histogram matching to human ground-truth maps prior to computing SIM and EMD
- note: old performances still available here
- added code for optimizing center, blur, and histogram matching to a human fixation distribution
- sample images added to page

18/09/2014
- code added for eDN

17/09/2014
- added Outlier Saliency (OS) model

13/08/2014
- added code to calculate degrees of visual angle

12/08/2014
- added 4 additional metrics (AUC-Borji, sAUC, CC, NSS) to results table
- created git repository to house matlab code
- added new 2014 TPAMI paper for Region Contrast model

11/08/2014
- added baseline 'one human' (i.e. 1 vs 38), to differentiate from baseline 'infinite' humans (i.e. inf vs inf)
- added Holzbach's "Sampled Template Collation" model

08/08/2014
- new model sent in by Shunji Satoh

07/08/2014
- model performances table is now sortable by different metrics
- added a new baseline: permutation control, based on Koehler et al. [JoV 2014]

29/07/2014
- two new saliency datasets have been included from Qi Zhao (corresponding to two ECCV 2014 papers)
- Matthieu Perreira Da Silva's Visual Conspicuity (VICO) code included

05/2014 - 07/2014
- all models have been retested, and latest timestamps included with results
AIM, GBVS, SUN, IttiKoch1, and IttiKoch2 performances have slightly improved since first tested (latest code used, download date included)

06/2014
- new website launched at http://saliency.mit.edu; added comparison table for other saliency datasets

01/2012
- website first launched at http://people.csail.mit.edu/tjudd/SaliencyBenchmark