In [1]:

    

projectDir = paths.concat(os.getenv('HOME'),'pose-hg-train')
paths.dofile(projectDir .. '/src/ref.lua')


    

    
Out[1]:





Saving everything to: /home/alejandro/pose-hg-train/exp/mpii/default	







    
Out[1]:





Input is a tensor with dimensions: 3 x 256 x 256	
Output is a table	
	 Entry 1 is a tensor with dimensions: 16 x 64 x 64	
	 Entry 2 is a tensor with dimensions: 16 x 64 x 64	
	 Entry 3 is a tensor with dimensions: 16 x 64 x 64	
	 Entry 4 is a tensor with dimensions: 16 x 64 x 64	
	 Entry 5 is a tensor with dimensions: 16 x 64 x 64	
	 Entry 6 is a tensor with dimensions: 16 x 64 x 64	
	 Entry 7 is a tensor with dimensions: 16 x 64 x 64	
	 Entry 8 is a tensor with dimensions: 16 x 64 x 64	

In [4]:

    

local set = 'train'
local tmpIdx = opt.idxRef[set][torch.random(dataset:size(set))]
print(tmpIdx)
inp,hms = heatmapVisualization(set,tmpIdx)
itorch.image(inp)
itorch.image(hms)


    

    
Out[4]:





20048	







    













    

In [14]:

    

idxs, preds, hms, inp = loadPreds('mpii/default/preds', true, false)


    

In [21]:

    

local tmpIdx = torch.random(idxs:size(1))
local img = dataset:loadImage(idxs[tmpIdx]) -- Load original image
drawSkeleton(img, preds[tmpIdx]:narrow(2,1,2):clone(), preds[tmpIdx]:narrow(2,5,1):clone():view(-1))

print("Predicted pose:"); sys.sleep(.01)
itorch.image(img); sys.sleep(.01)

if hms then
    -- Prepare heatmap visualization
    local hmImg = heatmapVisualization(nil,idxs[tmpIdx],hms[tmpIdx])
    print("Heatmaps: (red - ground truth, blue - predicted)"); sys.sleep(.01)
    itorch.image(hmImg)
end


    

    
Out[21]:





Predicted pose:	







    













    
Out[21]:





Heatmaps: (red - ground truth, blue - predicted)