Object Detection

Method VOC2007 VOC2010 VOC2012 ILSVRC 2013 MSCOCO 2015 Speed
OverFeat       24.3%    
R-CNN (AlexNet) 58.5% 53.7% 53.3% 31.4%    
R-CNN (VGG16) 66.0%          
SPP_net(ZF-5) 54.2%(1-model), 60.9%(2-model)     31.84%(1-model), 35.11%(6-model)    
DeepID-Net 64.1%     50.3%    
NoC 73.3%   68.8%      
Fast-RCNN (VGG16) 70.0% 68.8% 68.4%   19.7%(@[0.5-0.95]), 35.9%(@0.5)  
MR-CNN 78.2%   73.9%      
Faster-RCNN (VGG16) 78.8%   75.9%   21.9%(@[0.5-0.95]), 42.7%(@0.5) 198ms
Faster-RCNN (ResNet-101) 85.6%   83.8%   37.4%(@[0.5-0.95]), 59.0%(@0.5)  
SSD300 (VGG16) 77.2%   75.8%   25.1%(@[0.5-0.95]), 43.1%(@0.5) 46 fps
SSD512 (VGG16) 79.8%   78.5%   28.8%(@[0.5-0.95]), 48.5%(@0.5) 19 fps
ION 79.2%   76.4%      
CRAFT 75.7%   71.3% 48.5%    
OHEM 78.9%   76.3%   25.5%(@[0.5-0.95]), 45.9%(@0.5)  
R-FCN (ResNet-50) 77.4%         0.12sec(K40), 0.09sec(TitianX)
R-FCN (ResNet-101) 79.5%         0.17sec(K40), 0.12sec(TitianX)
R-FCN (ResNet-101),multi sc train 83.6%   82.0%   31.5%(@[0.5-0.95]), 53.2%(@0.5)  
PVANet 9.0 89.8%   84.2%     750ms(CPU), 46ms(TitianX)

Published: 09 Oct 2015

NLP

Tutorials

Published: 09 Oct 2015

Image Generation

Papers

Published: 09 Oct 2015

Generative Models and Generative Adversarial Network

Papers

Published: 09 Oct 2015

Fun With Deep Learning

Painting

Published: 09 Oct 2015

Deep Learning Tutorials

Tutorials

Published: 09 Oct 2015

Deep Learning Tricks

Papers

Published: 09 Oct 2015

Deep Learning Software and Hardware

Papers

Published: 09 Oct 2015