Classification Results: VOC2010 BETA

Competition "comp2" (train on own data)

This leaderboard shows only those submissions that have been marked as public, and so the displayed rankings should not be considered as definitive.

Average Precision (AP %)

  mean

aero
plane
bicycle

bird

boat

bottle

bus

car

cat

chair

cow

dining
table
dog

horse

motor
bike
person

potted
plant
sheep

sofa

train

tv/
monitor
submission
date
Improved Fisher Vector [?] 68.392.768.468.580.438.281.866.977.855.062.156.570.171.479.485.040.067.251.884.667.616-Aug-2010
Improved Fisher Vector [?] -92.768.069.079.929.381.460.078.045.062.931.669.271.278.678.034.067.3-82.7-10-Aug-2010
UCI_LSVM-MDPM-10X [?] --65.1---78.1---43.817.0-64.060.4--53.125.0-58.830-Aug-2010
LinearSVM-PHOW [?] 26.959.728.817.429.712.525.328.332.234.315.724.526.331.221.543.87.415.618.237.927.430-Aug-2010

Abbreviations

TitleMethodAffiliationContributorsDescriptionDate
Linear SVM on Improved Fisher vectorImproved Fisher VectorXRCEFlorent Perronnin Jorge Sanchez Thomas MensinkBased on [PSM10]: F. Perronnin, J. Sanchez and T. Mensink, "Improving the Fisher kernel for Large-Scale Image Classification", ECCV, 2010. Trained on close to 1M mono-tagged Flickr group images (non-overlapping with test set).2010-08-10 07:54:09
Linear SVM on Improved Fisher vectorImproved Fisher VectorXRCEFlorent Perronnin, Jorge Sanchez, Thomas MensinkBased on [PSM10]: F. Perronnin, J. Sanchez and T. Mensink, "Improving the Fisher kernel for Large-Scale Image Classification", ECCV, 2010. Late fusion of two systems trained respectively on i)voc10 trainval and ii) close to 1M mono-tagged Flickr group images. Optimal weights are learned per-class through cross-validation.2010-08-16 09:14:56
LinearSVM-PHOWLinearSVM-PHOWBeijing Institute of TechnologyChunliang Lv, Lu Tian, Yuan Zhou, Xiumin ShiLinear SVM classifier using spatial pyramid matching kernel. 2010-08-30 17:17:28
10x train set for LSVM, mixtures, deformable partsUCI_LSVM-MDPM-10XUniversity of California, IrvineXiangxin Zhu, Carl Vondrick, Deva Ramanan, Charless FowlkesWe downloaded additional images from Flickr that match the distribution of the testing set. We used Amazon's Mechanical Turk to annotate these training sets that are 10 times larger the standard trainval set. We used our larger training set to train models with the detector from Felzenswalb et. all.2010-08-30 04:33:45