PASCAL VOC Challenge performance evaluation server

Segmentation Results: VOC2010 ^BETA

Competition "comp6" (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.

The highest scoring entry in each column is shown in bold.
Clicking on the blue arrow symbol () at the top of a column will order the submissions from high to low wrt performance on that column.

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
	Oxford_TVG_CRF_RNN_COCO ^[?]	75.7	92.2	58.5	87.5	66.8	70.6	92.1	85.3	84.7	36.4	80.4	65.6	80.9	83.9	86.3	76.8	64.0	81.2	51.4	81.8	69.7	22-Apr-2015
	Oxford_TVG_CRF_RNN_VOC ^[?]	73.6	88.1	40.1	90.1	66.6	69.5	92.2	81.8	83.8	34.3	82.5	58.4	83.5	81.8	82.4	77.2	67.0	81.7	43.7	80.6	66.7	22-Apr-2015
	TTI_zoomout ^[?]	64.4	82.4	35.9	77.3	55.8	58.0	82.9	76.8	76.6	22.3	70.6	52.4	72.8	77.2	75.7	64.9	48.1	69.0	36.4	70.6	56.4	25-Nov-2014
	Berkeley-poselets-align-pb ^[?]	-	49.7	23.3	20.6	19.0	47.1	58.1	53.6	32.5	-	31.1	-	29.5	42.9	41.9	43.8	16.6	39.0	18.4	38.0	41.5	30-Aug-2010

Abbreviations

Title	Method	Affiliation	Contributors	Description	Date
Segmentation on poselet detections	Berkeley-poselets-align-pb	UC Berkeley	Thomas Brox, Lubomir Bourdev, Subhransu Maji, Jitendra Malik	Non-rigid alignment of poselet activations to UCM edges, filling of area between object edges by variational smoothing, competition among objects, and competitive refinement of masks	2010-08-30 21:00:35
Oxford_TVG_CRF_RNN_COCO	Oxford_TVG_CRF_RNN_COCO	[1] University of Oxford / [2] Baidu IDL	Shuai Zheng [1]; Sadeep Jayasumana [1]; Bernardino Romera-Paredes [1]; Chang Huang [2]; Philip Torr [1]	We introduce a new form of convolutional neural network, called CRF-RNN, which expresses Dense Conditional Random Fields (Dense CRF) as a Recurrent Neural Network (RNN). We plug this CRF-RNN network into an existing deep CNN to obtain a system that combines the desirable properties of both CNNs and CRFs. Importantly, our system fully integrates CRF inference with CNNs, making it possible to train the whole system end-to-end with the usual back-propagation algorithm. The system used for this submission was trained on VOC 2012 Segmentation challenge train data, Berkeley augmented data and a subset of COCO 2014 train data. More details will be available in the paper http://arxiv.org/abs/1502.03240.	2015-04-22 14:22:55
Oxford_TVG_CRF_RNN_VOC	Oxford_TVG_CRF_RNN_VOC	[1] University of Oxford / [2] Baidu IDL	Shuai Zheng [1]; Sadeep Jayasumana [1]; Bernardino Romera-Paredes [1]; Chang Huang [2]; Philip Torr [1]	We introduce a new form of convolutional neural network, called CRF-RNN, which expresses Dense Conditional Random Fields (Dense CRF) as a Recurrent Neural Network (RNN). We plug this CRF-RNN network into an existing deep CNN to obtain a system that combines the desirable properties of both CNNs and CRFs. Importantly, our system fully integrates CRF with CNNs, making it possible to train the whole system end-to-end with the usual back-propagation algorithm. The system used for this submission was trained on VOC 2012 Segmentation challenge train data, and Berkeley augmented data (COCO dataset was not used). More details will be available in the paper http://arxiv.org/abs/1502.03240.	2015-04-22 11:12:44
Feedforward segmentation with zoom-out features	TTI_zoomout	TTI-Chicago	Mohammadreza Mostajabi, Payman Yadollahpour, Gregory Shakhnarovich	Same as before, except using VGG 16-layer network instead of VGG CNN-S network. Fine-tuning on VOC-2012 was not performed. See http://arxiv.org/abs/1412.0774 for details.	2014-11-25 18:42:29