p04798526
/
OmniGibson
mirror of https://github.com/StanfordVL/OmniGibson.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

update show file

This commit is contained in:
fxia22 2017-09-05 13:19:30 -07:00
parent 1c3b83cde8
commit dff8d23ae5
1 changed files with 12 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
dev/show_3d2.py
View File

@ -75,8 +75,8 @@ def showpoints(imgs, depths, poses, model, target):
cv2.moveWindow('show3d',0,0)
cv2.setMouseCallback('show3d',onmouse)
imgv = Variable(torch.zeros(1,3, showsz, showsz*2)).cuda()
maskv = Variable(torch.zeros(1,1, showsz, showsz*2)).cuda()
imgv = Variable(torch.zeros(1,3, showsz, showsz*2), volatile=True).cuda()
maskv = Variable(torch.zeros(1,1, showsz, showsz*2), volatile=True).cuda()
cpose = np.eye(4)
@ -105,16 +105,21 @@ def showpoints(imgs, depths, poses, model, target):
if model:
tf = transforms.ToTensor()
source = tf(show)
source_depth = tf(np.expand_dims(target_depth, 2))
source_depth = tf(np.expand_dims(target_depth, 2).astype(np.float32)/65536 * 255)
#print(source.size(), source_depth.size())
imgv.data.copy_(source)
maskv.data.copy_(source_depth)
recon = model(imgv, maskv)
#print(recon.size())
recon = model(imgv[:,:,:1024, :1024], maskv[:,:,:1024, :1024])
print(recon.size())
show2 = recon.data.cpu().numpy()[0].transpose(1,2,0)
show[:] = (show2[:] * 255).astype(np.uint8)
show[:1024, :1024, :] = (show2[:] * 255).astype(np.uint8)
recon = model(imgv[:,:,:1024, 1024:], maskv[:,:,:1024, 1024:])
print(recon.size())
show2 = recon.data.cpu().numpy()[0].transpose(1,2,0)
show[:1024, 1024:, :] = (show2[:] * 255).astype(np.uint8)
t1 = time.time()
t = t1-t0
@ -237,7 +242,7 @@ if __name__=='__main__':
parser.add_argument('--model' , type = str, default = '', help='path of model')
opt = parser.parse_args()
d = ViewDataSet3D(root=opt.dataroot, transform = np.array, mist_transform = np.array, seqlen = 2, off_3d = False)
d = ViewDataSet3D(root=opt.dataroot, transform = np.array, mist_transform = np.array, seqlen = 5, off_3d = False)
idx = opt.idx
data = d[idx]