Giter VIP home page Giter VIP logo

generativepredictablevoxels's Introduction

Testing code release for

R. Girdhar, D. F. Fouhey, M. Rodriguez and A. Gupta
Learning a Predictable and Generative Vector Representation for Objects
In Proc. of European Conference on Computer Vision (ECCV), 2016

@inproceedings{Girdhar16b,
    title = {Learning a Predictable and Generative Vector Representation for Objects},
    author = {Girdhar, R. and Fouhey, D.F. and Rodriguez, M. and Gupta, A.},
    booktitle = {ECCV},
    year = {2016},
}

Pre-requisites

  1. Caffe (trained and tested with 97f4536, though should work with the latest version). Clone and install in libs dir.
  2. Python libs h5py, matplotlib, mayavi.

Download pre-trained models

Download all the models from here to models/ dir.

Testing using the precomputed networks

$ python src/testing/reconst.py  # stores the prediction in output/ folder

Data

The data was stored in HDF5 format for training. The total size of this set is quite large (around 0.5TB), which is hard to release, so I am sharing a subset of the data here.

The data can be accessed as follows (in python):

>>> import h5py
>>> f = h5py.File('batch_0.h5')
>>> images = f['data'].value; print(images.shape)
(198, 3, 227, 227)
>>> voxels = f['label-voxel'].value; voxels.shape
(198, 1, 20, 20, 20)

generativepredictablevoxels's People

Contributors

rohitgirdhar avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

generativepredictablevoxels's Issues

The link of Dropbox is invalid now

Hello, authors. I'm afraid that the link of Dropbox is invalid now. Is it convenient for you to check the link, or upload the required models into the models folder? Many thanks!
By the way, is there any useful tool to transform CAD model into voxel grids? You mentioned the voxelizer in your paper, however I cannot find it. Can you give me a hand to post the link. Thanks a lot.

Best wishes.

Cannot use legacy accessors on Blobs with > 4 axes

Hi, when running the reconst.py, I encounter this error after configuring the necessary libs.

Can you give me any suggestion on how can I fix it? Many thanks!

I1028 17:29:32.755017 3432 net.cpp:413] Input 0 -> data
I1028 17:29:32.755089 3432 layer_factory.hpp:77] Creating layer conv1
I1028 17:29:32.755129 3432 net.cpp:106] Creating Layer conv1
I1028 17:29:32.755147 3432 net.cpp:454] conv1 <- data
I1028 17:29:32.755167 3432 net.cpp:411] conv1 -> conv1
I1028 17:29:32.757350 3432 net.cpp:150] Setting up conv1
I1028 17:29:32.757392 3432 net.cpp:157] Top shape: 1 96 55 55 (290400)
I1028 17:29:32.757408 3432 net.cpp:165] Memory required for data: 1161600
I1028 17:29:32.757436 3432 layer_factory.hpp:77] Creating layer relu1
I1028 17:29:32.757458 3432 net.cpp:106] Creating Layer relu1
I1028 17:29:32.757474 3432 net.cpp:454] relu1 <- conv1
I1028 17:29:32.757491 3432 net.cpp:397] relu1 -> conv1 (in-place)
I1028 17:29:32.757513 3432 net.cpp:150] Setting up relu1
I1028 17:29:32.757529 3432 net.cpp:157] Top shape: 1 96 55 55 (290400)
I1028 17:29:32.757544 3432 net.cpp:165] Memory required for data: 2323200
I1028 17:29:32.757557 3432 layer_factory.hpp:77] Creating layer pool1
I1028 17:29:32.757575 3432 net.cpp:106] Creating Layer pool1
I1028 17:29:32.757591 3432 net.cpp:454] pool1 <- conv1
I1028 17:29:32.757609 3432 net.cpp:411] pool1 -> pool1
I1028 17:29:32.757637 3432 net.cpp:150] Setting up pool1
I1028 17:29:32.757657 3432 net.cpp:157] Top shape: 1 96 27 27 (69984)
I1028 17:29:32.757669 3432 net.cpp:165] Memory required for data: 2603136
I1028 17:29:32.757683 3432 layer_factory.hpp:77] Creating layer norm1
I1028 17:29:32.757704 3432 net.cpp:106] Creating Layer norm1
I1028 17:29:32.757719 3432 net.cpp:454] norm1 <- pool1
I1028 17:29:32.757735 3432 net.cpp:411] norm1 -> norm1
I1028 17:29:32.757761 3432 net.cpp:150] Setting up norm1
I1028 17:29:32.757777 3432 net.cpp:157] Top shape: 1 96 27 27 (69984)
I1028 17:29:32.757791 3432 net.cpp:165] Memory required for data: 2883072
I1028 17:29:32.757803 3432 layer_factory.hpp:77] Creating layer conv2
I1028 17:29:32.757829 3432 net.cpp:106] Creating Layer conv2
I1028 17:29:32.757845 3432 net.cpp:454] conv2 <- norm1
I1028 17:29:32.757879 3432 net.cpp:411] conv2 -> conv2
I1028 17:29:32.774193 3432 net.cpp:150] Setting up conv2
I1028 17:29:32.774226 3432 net.cpp:157] Top shape: 1 256 27 27 (186624)
I1028 17:29:32.774241 3432 net.cpp:165] Memory required for data: 3629568
I1028 17:29:32.774263 3432 layer_factory.hpp:77] Creating layer relu2
I1028 17:29:32.774283 3432 net.cpp:106] Creating Layer relu2
I1028 17:29:32.774302 3432 net.cpp:454] relu2 <- conv2
I1028 17:29:32.774319 3432 net.cpp:397] relu2 -> conv2 (in-place)
I1028 17:29:32.774340 3432 net.cpp:150] Setting up relu2
I1028 17:29:32.774358 3432 net.cpp:157] Top shape: 1 256 27 27 (186624)
I1028 17:29:32.774374 3432 net.cpp:165] Memory required for data: 4376064
I1028 17:29:32.774389 3432 layer_factory.hpp:77] Creating layer pool2
I1028 17:29:32.774410 3432 net.cpp:106] Creating Layer pool2
I1028 17:29:32.774425 3432 net.cpp:454] pool2 <- conv2
I1028 17:29:32.774443 3432 net.cpp:411] pool2 -> pool2
I1028 17:29:32.774467 3432 net.cpp:150] Setting up pool2
I1028 17:29:32.774487 3432 net.cpp:157] Top shape: 1 256 13 13 (43264)
I1028 17:29:32.774500 3432 net.cpp:165] Memory required for data: 4549120
I1028 17:29:32.774515 3432 layer_factory.hpp:77] Creating layer norm2
I1028 17:29:32.774536 3432 net.cpp:106] Creating Layer norm2
I1028 17:29:32.774551 3432 net.cpp:454] norm2 <- pool2
I1028 17:29:32.774569 3432 net.cpp:411] norm2 -> norm2
I1028 17:29:32.774591 3432 net.cpp:150] Setting up norm2
I1028 17:29:32.774610 3432 net.cpp:157] Top shape: 1 256 13 13 (43264)
I1028 17:29:32.774624 3432 net.cpp:165] Memory required for data: 4722176
I1028 17:29:32.774639 3432 layer_factory.hpp:77] Creating layer conv3
I1028 17:29:32.774667 3432 net.cpp:106] Creating Layer conv3
I1028 17:29:32.774684 3432 net.cpp:454] conv3 <- norm2
I1028 17:29:32.774703 3432 net.cpp:411] conv3 -> conv3
I1028 17:29:32.816196 3432 net.cpp:150] Setting up conv3
I1028 17:29:32.816229 3432 net.cpp:157] Top shape: 1 384 13 13 (64896)
I1028 17:29:32.816244 3432 net.cpp:165] Memory required for data: 4981760
I1028 17:29:32.816267 3432 layer_factory.hpp:77] Creating layer relu3
I1028 17:29:32.816285 3432 net.cpp:106] Creating Layer relu3
I1028 17:29:32.816301 3432 net.cpp:454] relu3 <- conv3
I1028 17:29:32.816318 3432 net.cpp:397] relu3 -> conv3 (in-place)
I1028 17:29:32.816339 3432 net.cpp:150] Setting up relu3
I1028 17:29:32.816357 3432 net.cpp:157] Top shape: 1 384 13 13 (64896)
I1028 17:29:32.816372 3432 net.cpp:165] Memory required for data: 5241344
I1028 17:29:32.816387 3432 layer_factory.hpp:77] Creating layer conv4
I1028 17:29:32.816411 3432 net.cpp:106] Creating Layer conv4
I1028 17:29:32.816428 3432 net.cpp:454] conv4 <- conv3
I1028 17:29:32.816448 3432 net.cpp:411] conv4 -> conv4
I1028 17:29:32.847182 3432 net.cpp:150] Setting up conv4
I1028 17:29:32.847221 3432 net.cpp:157] Top shape: 1 384 13 13 (64896)
I1028 17:29:32.847236 3432 net.cpp:165] Memory required for data: 5500928
I1028 17:29:32.847255 3432 layer_factory.hpp:77] Creating layer relu4
I1028 17:29:32.847275 3432 net.cpp:106] Creating Layer relu4
I1028 17:29:32.847295 3432 net.cpp:454] relu4 <- conv4
I1028 17:29:32.847316 3432 net.cpp:397] relu4 -> conv4 (in-place)
I1028 17:29:32.847337 3432 net.cpp:150] Setting up relu4
I1028 17:29:32.847357 3432 net.cpp:157] Top shape: 1 384 13 13 (64896)
I1028 17:29:32.847371 3432 net.cpp:165] Memory required for data: 5760512
I1028 17:29:32.847388 3432 layer_factory.hpp:77] Creating layer conv5
I1028 17:29:32.847409 3432 net.cpp:106] Creating Layer conv5
I1028 17:29:32.847425 3432 net.cpp:454] conv5 <- conv4
I1028 17:29:32.847445 3432 net.cpp:411] conv5 -> conv5
I1028 17:29:32.868224 3432 net.cpp:150] Setting up conv5
I1028 17:29:32.868255 3432 net.cpp:157] Top shape: 1 256 13 13 (43264)
I1028 17:29:32.868270 3432 net.cpp:165] Memory required for data: 5933568
I1028 17:29:32.868299 3432 layer_factory.hpp:77] Creating layer relu5
I1028 17:29:32.868320 3432 net.cpp:106] Creating Layer relu5
I1028 17:29:32.868336 3432 net.cpp:454] relu5 <- conv5
I1028 17:29:32.868355 3432 net.cpp:397] relu5 -> conv5 (in-place)
I1028 17:29:32.868376 3432 net.cpp:150] Setting up relu5
I1028 17:29:32.868392 3432 net.cpp:157] Top shape: 1 256 13 13 (43264)
I1028 17:29:32.868407 3432 net.cpp:165] Memory required for data: 6106624
I1028 17:29:32.868422 3432 layer_factory.hpp:77] Creating layer pool5
I1028 17:29:32.868444 3432 net.cpp:106] Creating Layer pool5
I1028 17:29:32.868459 3432 net.cpp:454] pool5 <- conv5
I1028 17:29:32.868479 3432 net.cpp:411] pool5 -> pool5
I1028 17:29:32.868504 3432 net.cpp:150] Setting up pool5
I1028 17:29:32.868522 3432 net.cpp:157] Top shape: 1 256 6 6 (9216)
I1028 17:29:32.868540 3432 net.cpp:165] Memory required for data: 6143488
I1028 17:29:32.868554 3432 layer_factory.hpp:77] Creating layer fc6
I1028 17:29:32.868579 3432 net.cpp:106] Creating Layer fc6
I1028 17:29:32.868615 3432 net.cpp:454] fc6 <- pool5
I1028 17:29:32.868638 3432 net.cpp:411] fc6 -> fc6
I1028 17:29:34.558751 3432 net.cpp:150] Setting up fc6
I1028 17:29:34.558848 3432 net.cpp:157] Top shape: 1 4096 (4096)
I1028 17:29:34.558862 3432 net.cpp:165] Memory required for data: 6159872
I1028 17:29:34.558899 3432 layer_factory.hpp:77] Creating layer relu6
I1028 17:29:34.558953 3432 net.cpp:106] Creating Layer relu6
I1028 17:29:34.558979 3432 net.cpp:454] relu6 <- fc6
I1028 17:29:34.558998 3432 net.cpp:397] relu6 -> fc6 (in-place)
I1028 17:29:34.559021 3432 net.cpp:150] Setting up relu6
I1028 17:29:34.559036 3432 net.cpp:157] Top shape: 1 4096 (4096)
I1028 17:29:34.559047 3432 net.cpp:165] Memory required for data: 6176256
I1028 17:29:34.559062 3432 layer_factory.hpp:77] Creating layer drop6
I1028 17:29:34.559089 3432 net.cpp:106] Creating Layer drop6
I1028 17:29:34.559103 3432 net.cpp:454] drop6 <- fc6
I1028 17:29:34.559120 3432 net.cpp:397] drop6 -> fc6 (in-place)
I1028 17:29:34.559140 3432 net.cpp:150] Setting up drop6
I1028 17:29:34.559155 3432 net.cpp:157] Top shape: 1 4096 (4096)
I1028 17:29:34.559167 3432 net.cpp:165] Memory required for data: 6192640
I1028 17:29:34.559180 3432 layer_factory.hpp:77] Creating layer fc7
I1028 17:29:34.559200 3432 net.cpp:106] Creating Layer fc7
I1028 17:29:34.559213 3432 net.cpp:454] fc7 <- fc6
I1028 17:29:34.559237 3432 net.cpp:411] fc7 -> fc7
I1028 17:29:35.191236 3432 net.cpp:150] Setting up fc7
I1028 17:29:35.191305 3432 net.cpp:157] Top shape: 1 4096 (4096)
I1028 17:29:35.191318 3432 net.cpp:165] Memory required for data: 6209024
I1028 17:29:35.191344 3432 layer_factory.hpp:77] Creating layer relu7
I1028 17:29:35.191366 3432 net.cpp:106] Creating Layer relu7
I1028 17:29:35.191385 3432 net.cpp:454] relu7 <- fc7
I1028 17:29:35.191404 3432 net.cpp:397] relu7 -> fc7 (in-place)
I1028 17:29:35.191426 3432 net.cpp:150] Setting up relu7
I1028 17:29:35.191442 3432 net.cpp:157] Top shape: 1 4096 (4096)
I1028 17:29:35.191453 3432 net.cpp:165] Memory required for data: 6225408
I1028 17:29:35.191468 3432 layer_factory.hpp:77] Creating layer drop7
I1028 17:29:35.191485 3432 net.cpp:106] Creating Layer drop7
I1028 17:29:35.191498 3432 net.cpp:454] drop7 <- fc7
I1028 17:29:35.191515 3432 net.cpp:397] drop7 -> fc7 (in-place)
I1028 17:29:35.191534 3432 net.cpp:150] Setting up drop7
I1028 17:29:35.191548 3432 net.cpp:157] Top shape: 1 4096 (4096)
I1028 17:29:35.191560 3432 net.cpp:165] Memory required for data: 6241792
I1028 17:29:35.191573 3432 layer_factory.hpp:77] Creating layer fc8-imfeat
I1028 17:29:35.191593 3432 net.cpp:106] Creating Layer fc8-imfeat
I1028 17:29:35.191606 3432 net.cpp:454] fc8-imfeat <- fc7
I1028 17:29:35.191624 3432 net.cpp:411] fc8-imfeat -> fc8-imfeat
I1028 17:29:35.201620 3432 net.cpp:150] Setting up fc8-imfeat
I1028 17:29:35.201645 3432 net.cpp:157] Top shape: 1 64 (64)
I1028 17:29:35.201658 3432 net.cpp:165] Memory required for data: 6242048
I1028 17:29:35.201673 3432 layer_factory.hpp:77] Creating layer small-2
I1028 17:29:35.201689 3432 net.cpp:106] Creating Layer small-2
I1028 17:29:35.201704 3432 net.cpp:454] small-2 <- fc8-imfeat
I1028 17:29:35.201719 3432 net.cpp:411] small-2 -> small-2
I1028 17:29:35.202283 3432 net.cpp:150] Setting up small-2
I1028 17:29:35.202304 3432 net.cpp:157] Top shape: 1 216 (216)
I1028 17:29:35.202316 3432 net.cpp:165] Memory required for data: 6242912
I1028 17:29:35.202337 3432 layer_factory.hpp:77] Creating layer reshape
I1028 17:29:35.202359 3432 net.cpp:106] Creating Layer reshape
I1028 17:29:35.202373 3432 net.cpp:454] reshape <- small-2
I1028 17:29:35.202386 3432 net.cpp:411] reshape -> small-reshaped
I1028 17:29:35.202409 3432 net.cpp:150] Setting up reshape
I1028 17:29:35.202422 3432 net.cpp:157] Top shape: 1 1 6 6 6 (216)
I1028 17:29:35.202443 3432 net.cpp:165] Memory required for data: 6243776
I1028 17:29:35.202456 3432 layer_factory.hpp:77] Creating layer deconv4_3d
I1028 17:29:35.202478 3432 net.cpp:106] Creating Layer deconv4_3d
I1028 17:29:35.202491 3432 net.cpp:454] deconv4_3d <- small-reshaped
I1028 17:29:35.202508 3432 net.cpp:411] deconv4_3d -> deconv4_3d
I1028 17:29:35.202813 3432 net.cpp:150] Setting up deconv4_3d
I1028 17:29:35.202834 3432 net.cpp:157] Top shape: 1 256 8 8 8 (131072)
I1028 17:29:35.202847 3432 net.cpp:165] Memory required for data: 6768064
I1028 17:29:35.202862 3432 layer_factory.hpp:77] Creating layer derelu4_3d
I1028 17:29:35.202879 3432 net.cpp:106] Creating Layer derelu4_3d
I1028 17:29:35.202891 3432 net.cpp:454] derelu4_3d <- deconv4_3d
I1028 17:29:35.202905 3432 net.cpp:397] derelu4_3d -> deconv4_3d (in-place)
F1028 17:29:35.202927 3432 blob.hpp:140] Check failed: num_axes() <= 4 (5 vs. 4) Cannot use legacy accessors on Blobs with > 4 axes.
*** Check failure stack trace: ***
Aborted (core dumped)

X Error of failed request: BadValue (integer parameter out of range for operation)

I faced the same issue stated in Issue #1 and followed the solution there to solve the problem. It does work to avoid the error message, but I am currently facing another one. Not sure if it is caused by commenting the check lines. Could you please give me suggestion on how to fix it?

I0310 00:29:05.668407  3796 layer_factory.hpp:77] Creating layer input
I0310 00:29:05.668440  3796 net.cpp:84] Creating Layer input
I0310 00:29:05.668452  3796 net.cpp:380] input -> data
I0310 00:29:05.668478  3796 net.cpp:122] Setting up input
I0310 00:29:05.708513  3796 net.cpp:129] Top shape: 1 3 227 227 (154587)
I0310 00:29:05.708537  3796 net.cpp:137] Memory required for data: 618348
I0310 00:29:05.708549  3796 layer_factory.hpp:77] Creating layer conv1
I0310 00:29:05.708575  3796 net.cpp:84] Creating Layer conv1
I0310 00:29:05.708581  3796 net.cpp:406] conv1 <- data
I0310 00:29:05.708592  3796 net.cpp:380] conv1 -> conv1
I0310 00:29:05.790503  3796 net.cpp:122] Setting up conv1
I0310 00:29:05.790529  3796 net.cpp:129] Top shape: 1 96 55 55 (290400)
I0310 00:29:05.790534  3796 net.cpp:137] Memory required for data: 1779948
I0310 00:29:05.790546  3796 layer_factory.hpp:77] Creating layer relu1
I0310 00:29:05.790558  3796 net.cpp:84] Creating Layer relu1
I0310 00:29:05.790562  3796 net.cpp:406] relu1 <- conv1
I0310 00:29:05.790568  3796 net.cpp:367] relu1 -> conv1 (in-place)
I0310 00:29:05.790576  3796 net.cpp:122] Setting up relu1
I0310 00:29:05.790581  3796 net.cpp:129] Top shape: 1 96 55 55 (290400)
I0310 00:29:05.790585  3796 net.cpp:137] Memory required for data: 2941548
I0310 00:29:05.790588  3796 layer_factory.hpp:77] Creating layer pool1
I0310 00:29:05.790596  3796 net.cpp:84] Creating Layer pool1
I0310 00:29:05.790599  3796 net.cpp:406] pool1 <- conv1
I0310 00:29:05.790603  3796 net.cpp:380] pool1 -> pool1
I0310 00:29:05.812721  3796 net.cpp:122] Setting up pool1
I0310 00:29:05.812747  3796 net.cpp:129] Top shape: 1 96 27 27 (69984)
I0310 00:29:05.812753  3796 net.cpp:137] Memory required for data: 3221484
I0310 00:29:05.812759  3796 layer_factory.hpp:77] Creating layer norm1
I0310 00:29:05.812774  3796 net.cpp:84] Creating Layer norm1
I0310 00:29:05.812780  3796 net.cpp:406] norm1 <- pool1
I0310 00:29:05.812788  3796 net.cpp:380] norm1 -> norm1
I0310 00:29:05.813802  3796 net.cpp:122] Setting up norm1
I0310 00:29:05.813813  3796 net.cpp:129] Top shape: 1 96 27 27 (69984)
I0310 00:29:05.813818  3796 net.cpp:137] Memory required for data: 3501420
I0310 00:29:05.813824  3796 layer_factory.hpp:77] Creating layer conv2
I0310 00:29:05.813840  3796 net.cpp:84] Creating Layer conv2
I0310 00:29:05.813846  3796 net.cpp:406] conv2 <- norm1
I0310 00:29:05.813854  3796 net.cpp:380] conv2 -> conv2
I0310 00:29:05.818429  3796 net.cpp:122] Setting up conv2
I0310 00:29:05.818441  3796 net.cpp:129] Top shape: 1 256 27 27 (186624)
I0310 00:29:05.818446  3796 net.cpp:137] Memory required for data: 4247916
I0310 00:29:05.818459  3796 layer_factory.hpp:77] Creating layer relu2
I0310 00:29:05.818469  3796 net.cpp:84] Creating Layer relu2
I0310 00:29:05.818473  3796 net.cpp:406] relu2 <- conv2
I0310 00:29:05.818482  3796 net.cpp:367] relu2 -> conv2 (in-place)
I0310 00:29:05.818491  3796 net.cpp:122] Setting up relu2
I0310 00:29:05.818497  3796 net.cpp:129] Top shape: 1 256 27 27 (186624)
I0310 00:29:05.818502  3796 net.cpp:137] Memory required for data: 4994412
I0310 00:29:05.818507  3796 layer_factory.hpp:77] Creating layer pool2
I0310 00:29:05.818516  3796 net.cpp:84] Creating Layer pool2
I0310 00:29:05.818521  3796 net.cpp:406] pool2 <- conv2
I0310 00:29:05.818536  3796 net.cpp:380] pool2 -> pool2
I0310 00:29:05.818549  3796 net.cpp:122] Setting up pool2
I0310 00:29:05.818557  3796 net.cpp:129] Top shape: 1 256 13 13 (43264)
I0310 00:29:05.818562  3796 net.cpp:137] Memory required for data: 5167468
I0310 00:29:05.818567  3796 layer_factory.hpp:77] Creating layer norm2
I0310 00:29:05.818577  3796 net.cpp:84] Creating Layer norm2
I0310 00:29:05.818581  3796 net.cpp:406] norm2 <- pool2
I0310 00:29:05.818588  3796 net.cpp:380] norm2 -> norm2
I0310 00:29:05.818598  3796 net.cpp:122] Setting up norm2
I0310 00:29:05.818616  3796 net.cpp:129] Top shape: 1 256 13 13 (43264)
I0310 00:29:05.818621  3796 net.cpp:137] Memory required for data: 5340524
I0310 00:29:05.818626  3796 layer_factory.hpp:77] Creating layer conv3
I0310 00:29:05.818639  3796 net.cpp:84] Creating Layer conv3
I0310 00:29:05.818644  3796 net.cpp:406] conv3 <- norm2
I0310 00:29:05.818651  3796 net.cpp:380] conv3 -> conv3
I0310 00:29:05.831708  3796 net.cpp:122] Setting up conv3
I0310 00:29:05.831734  3796 net.cpp:129] Top shape: 1 384 13 13 (64896)
I0310 00:29:05.831740  3796 net.cpp:137] Memory required for data: 5600108
I0310 00:29:05.831754  3796 layer_factory.hpp:77] Creating layer relu3
I0310 00:29:05.831765  3796 net.cpp:84] Creating Layer relu3
I0310 00:29:05.831771  3796 net.cpp:406] relu3 <- conv3
I0310 00:29:05.831779  3796 net.cpp:367] relu3 -> conv3 (in-place)
I0310 00:29:05.831789  3796 net.cpp:122] Setting up relu3
I0310 00:29:05.831794  3796 net.cpp:129] Top shape: 1 384 13 13 (64896)
I0310 00:29:05.831799  3796 net.cpp:137] Memory required for data: 5859692
I0310 00:29:05.831804  3796 layer_factory.hpp:77] Creating layer conv4
I0310 00:29:05.831817  3796 net.cpp:84] Creating Layer conv4
I0310 00:29:05.831823  3796 net.cpp:406] conv4 <- conv3
I0310 00:29:05.831832  3796 net.cpp:380] conv4 -> conv4
I0310 00:29:05.841760  3796 net.cpp:122] Setting up conv4
I0310 00:29:05.841790  3796 net.cpp:129] Top shape: 1 384 13 13 (64896)
I0310 00:29:05.841796  3796 net.cpp:137] Memory required for data: 6119276
I0310 00:29:05.841809  3796 layer_factory.hpp:77] Creating layer relu4
I0310 00:29:05.841822  3796 net.cpp:84] Creating Layer relu4
I0310 00:29:05.841828  3796 net.cpp:406] relu4 <- conv4
I0310 00:29:05.841837  3796 net.cpp:367] relu4 -> conv4 (in-place)
I0310 00:29:05.841848  3796 net.cpp:122] Setting up relu4
I0310 00:29:05.841856  3796 net.cpp:129] Top shape: 1 384 13 13 (64896)
I0310 00:29:05.841861  3796 net.cpp:137] Memory required for data: 6378860
I0310 00:29:05.841864  3796 layer_factory.hpp:77] Creating layer conv5
I0310 00:29:05.841879  3796 net.cpp:84] Creating Layer conv5
I0310 00:29:05.841883  3796 net.cpp:406] conv5 <- conv4
I0310 00:29:05.841893  3796 net.cpp:380] conv5 -> conv5
I0310 00:29:05.848516  3796 net.cpp:122] Setting up conv5
I0310 00:29:05.848546  3796 net.cpp:129] Top shape: 1 256 13 13 (43264)
I0310 00:29:05.848552  3796 net.cpp:137] Memory required for data: 6551916
I0310 00:29:05.848570  3796 layer_factory.hpp:77] Creating layer relu5
I0310 00:29:05.848583  3796 net.cpp:84] Creating Layer relu5
I0310 00:29:05.848590  3796 net.cpp:406] relu5 <- conv5
I0310 00:29:05.848599  3796 net.cpp:367] relu5 -> conv5 (in-place)
I0310 00:29:05.848610  3796 net.cpp:122] Setting up relu5
I0310 00:29:05.848618  3796 net.cpp:129] Top shape: 1 256 13 13 (43264)
I0310 00:29:05.848623  3796 net.cpp:137] Memory required for data: 6724972
I0310 00:29:05.848628  3796 layer_factory.hpp:77] Creating layer pool5
I0310 00:29:05.848637  3796 net.cpp:84] Creating Layer pool5
I0310 00:29:05.848641  3796 net.cpp:406] pool5 <- conv5
I0310 00:29:05.848650  3796 net.cpp:380] pool5 -> pool5
I0310 00:29:05.848662  3796 net.cpp:122] Setting up pool5
I0310 00:29:05.848668  3796 net.cpp:129] Top shape: 1 256 6 6 (9216)
I0310 00:29:05.848673  3796 net.cpp:137] Memory required for data: 6761836
I0310 00:29:05.848678  3796 layer_factory.hpp:77] Creating layer fc6
I0310 00:29:05.848692  3796 net.cpp:84] Creating Layer fc6
I0310 00:29:05.848696  3796 net.cpp:406] fc6 <- pool5
I0310 00:29:05.848706  3796 net.cpp:380] fc6 -> fc6
I0310 00:29:06.367887  3796 net.cpp:122] Setting up fc6
I0310 00:29:06.367926  3796 net.cpp:129] Top shape: 1 4096 (4096)
I0310 00:29:06.367931  3796 net.cpp:137] Memory required for data: 6778220
I0310 00:29:06.367949  3796 layer_factory.hpp:77] Creating layer relu6
I0310 00:29:06.367964  3796 net.cpp:84] Creating Layer relu6
I0310 00:29:06.367974  3796 net.cpp:406] relu6 <- fc6
I0310 00:29:06.367983  3796 net.cpp:367] relu6 -> fc6 (in-place)
I0310 00:29:06.367995  3796 net.cpp:122] Setting up relu6
I0310 00:29:06.368002  3796 net.cpp:129] Top shape: 1 4096 (4096)
I0310 00:29:06.368007  3796 net.cpp:137] Memory required for data: 6794604
I0310 00:29:06.368011  3796 layer_factory.hpp:77] Creating layer drop6
I0310 00:29:06.368022  3796 net.cpp:84] Creating Layer drop6
I0310 00:29:06.368027  3796 net.cpp:406] drop6 <- fc6
I0310 00:29:06.368036  3796 net.cpp:367] drop6 -> fc6 (in-place)
I0310 00:29:06.368046  3796 net.cpp:122] Setting up drop6
I0310 00:29:06.368052  3796 net.cpp:129] Top shape: 1 4096 (4096)
I0310 00:29:06.368057  3796 net.cpp:137] Memory required for data: 6810988
I0310 00:29:06.368062  3796 layer_factory.hpp:77] Creating layer fc7
I0310 00:29:06.368072  3796 net.cpp:84] Creating Layer fc7
I0310 00:29:06.368077  3796 net.cpp:406] fc7 <- fc6
I0310 00:29:06.368085  3796 net.cpp:380] fc7 -> fc7
I0310 00:29:06.556226  3796 net.cpp:122] Setting up fc7
I0310 00:29:06.556262  3796 net.cpp:129] Top shape: 1 4096 (4096)
I0310 00:29:06.556277  3796 net.cpp:137] Memory required for data: 6827372
I0310 00:29:06.556293  3796 layer_factory.hpp:77] Creating layer relu7
I0310 00:29:06.556313  3796 net.cpp:84] Creating Layer relu7
I0310 00:29:06.556323  3796 net.cpp:406] relu7 <- fc7
I0310 00:29:06.556334  3796 net.cpp:367] relu7 -> fc7 (in-place)
I0310 00:29:06.556347  3796 net.cpp:122] Setting up relu7
I0310 00:29:06.556355  3796 net.cpp:129] Top shape: 1 4096 (4096)
I0310 00:29:06.556362  3796 net.cpp:137] Memory required for data: 6843756
I0310 00:29:06.556370  3796 layer_factory.hpp:77] Creating layer drop7
I0310 00:29:06.556381  3796 net.cpp:84] Creating Layer drop7
I0310 00:29:06.556388  3796 net.cpp:406] drop7 <- fc7
I0310 00:29:06.556398  3796 net.cpp:367] drop7 -> fc7 (in-place)
I0310 00:29:06.556411  3796 net.cpp:122] Setting up drop7
I0310 00:29:06.556418  3796 net.cpp:129] Top shape: 1 4096 (4096)
I0310 00:29:06.556426  3796 net.cpp:137] Memory required for data: 6860140
I0310 00:29:06.556433  3796 layer_factory.hpp:77] Creating layer fc8-imfeat
I0310 00:29:06.556445  3796 net.cpp:84] Creating Layer fc8-imfeat
I0310 00:29:06.556453  3796 net.cpp:406] fc8-imfeat <- fc7
I0310 00:29:06.556465  3796 net.cpp:380] fc8-imfeat -> fc8-imfeat
I0310 00:29:06.560536  3796 net.cpp:122] Setting up fc8-imfeat
I0310 00:29:06.560576  3796 net.cpp:129] Top shape: 1 64 (64)
I0310 00:29:06.560585  3796 net.cpp:137] Memory required for data: 6860396
I0310 00:29:06.560601  3796 layer_factory.hpp:77] Creating layer small-2
I0310 00:29:06.560627  3796 net.cpp:84] Creating Layer small-2
I0310 00:29:06.560642  3796 net.cpp:406] small-2 <- fc8-imfeat
I0310 00:29:06.560663  3796 net.cpp:380] small-2 -> small-2
I0310 00:29:06.560907  3796 net.cpp:122] Setting up small-2
I0310 00:29:06.560921  3796 net.cpp:129] Top shape: 1 216 (216)
I0310 00:29:06.560941  3796 net.cpp:137] Memory required for data: 6861260
I0310 00:29:06.560962  3796 layer_factory.hpp:77] Creating layer reshape
I0310 00:29:06.560977  3796 net.cpp:84] Creating Layer reshape
I0310 00:29:06.560986  3796 net.cpp:406] reshape <- small-2
I0310 00:29:06.560994  3796 net.cpp:380] reshape -> small-reshaped
I0310 00:29:06.561015  3796 net.cpp:122] Setting up reshape
I0310 00:29:06.561025  3796 net.cpp:129] Top shape: 1 1 6 6 6 (216)
I0310 00:29:06.561038  3796 net.cpp:137] Memory required for data: 6862124
I0310 00:29:06.561045  3796 layer_factory.hpp:77] Creating layer deconv4_3d
I0310 00:29:06.561064  3796 net.cpp:84] Creating Layer deconv4_3d
I0310 00:29:06.561082  3796 net.cpp:406] deconv4_3d <- small-reshaped
I0310 00:29:06.561094  3796 net.cpp:380] deconv4_3d -> deconv4_3d
I0310 00:29:06.561246  3796 net.cpp:122] Setting up deconv4_3d
I0310 00:29:06.561261  3796 net.cpp:129] Top shape: 1 256 8 8 8 (131072)
I0310 00:29:06.561269  3796 net.cpp:137] Memory required for data: 7386412
I0310 00:29:06.561280  3796 layer_factory.hpp:77] Creating layer derelu4_3d
I0310 00:29:06.561291  3796 net.cpp:84] Creating Layer derelu4_3d
I0310 00:29:06.561298  3796 net.cpp:406] derelu4_3d <- deconv4_3d
I0310 00:29:06.561318  3796 net.cpp:367] derelu4_3d -> deconv4_3d (in-place)
I0310 00:29:06.561628  3796 net.cpp:122] Setting up derelu4_3d
I0310 00:29:06.561645  3796 net.cpp:129] Top shape: 1 256 8 8 8 (131072)
I0310 00:29:06.561652  3796 net.cpp:137] Memory required for data: 7910700
I0310 00:29:06.561662  3796 layer_factory.hpp:77] Creating layer deconv3_3d
I0310 00:29:06.561676  3796 net.cpp:84] Creating Layer deconv3_3d
I0310 00:29:06.561683  3796 net.cpp:406] deconv3_3d <- deconv4_3d
I0310 00:29:06.561693  3796 net.cpp:380] deconv3_3d -> deconv3_3d
I0310 00:29:06.590132  3796 net.cpp:122] Setting up deconv3_3d
I0310 00:29:06.590168  3796 net.cpp:129] Top shape: 1 384 10 10 10 (384000)
I0310 00:29:06.590175  3796 net.cpp:137] Memory required for data: 9446700
I0310 00:29:06.590188  3796 layer_factory.hpp:77] Creating layer derelu3_3d
I0310 00:29:06.590204  3796 net.cpp:84] Creating Layer derelu3_3d
I0310 00:29:06.590211  3796 net.cpp:406] derelu3_3d <- deconv3_3d
I0310 00:29:06.590220  3796 net.cpp:367] derelu3_3d -> deconv3_3d (in-place)
I0310 00:29:06.591117  3796 net.cpp:122] Setting up derelu3_3d
I0310 00:29:06.591140  3796 net.cpp:129] Top shape: 1 384 10 10 10 (384000)
I0310 00:29:06.591154  3796 net.cpp:137] Memory required for data: 10982700
I0310 00:29:06.591166  3796 layer_factory.hpp:77] Creating layer deconv2_3d
I0310 00:29:06.591183  3796 net.cpp:84] Creating Layer deconv2_3d
I0310 00:29:06.591192  3796 net.cpp:406] deconv2_3d <- deconv3_3d
I0310 00:29:06.591204  3796 net.cpp:380] deconv2_3d -> deconv2_3d
I0310 00:29:06.730216  3796 net.cpp:122] Setting up deconv2_3d
I0310 00:29:06.730244  3796 net.cpp:129] Top shape: 1 256 14 14 14 (702464)
I0310 00:29:06.730248  3796 net.cpp:137] Memory required for data: 13792556
I0310 00:29:06.730258  3796 layer_factory.hpp:77] Creating layer derelu2_3d
I0310 00:29:06.730270  3796 net.cpp:84] Creating Layer derelu2_3d
I0310 00:29:06.730274  3796 net.cpp:406] derelu2_3d <- deconv2_3d
I0310 00:29:06.730281  3796 net.cpp:367] derelu2_3d -> deconv2_3d (in-place)
I0310 00:29:06.731506  3796 net.cpp:122] Setting up derelu2_3d
I0310 00:29:06.731519  3796 net.cpp:129] Top shape: 1 256 14 14 14 (702464)
I0310 00:29:06.731524  3796 net.cpp:137] Memory required for data: 16602412
I0310 00:29:06.731528  3796 layer_factory.hpp:77] Creating layer deconv1_3d
I0310 00:29:06.731537  3796 net.cpp:84] Creating Layer deconv1_3d
I0310 00:29:06.731541  3796 net.cpp:406] deconv1_3d <- deconv2_3d
I0310 00:29:06.731549  3796 net.cpp:380] deconv1_3d -> deconv1_3d
I0310 00:29:06.824317  3796 net.cpp:122] Setting up deconv1_3d
I0310 00:29:06.824348  3796 net.cpp:129] Top shape: 1 96 20 20 20 (768000)
I0310 00:29:06.824352  3796 net.cpp:137] Memory required for data: 19674412
I0310 00:29:06.824362  3796 layer_factory.hpp:77] Creating layer derelu1_3d
I0310 00:29:06.824378  3796 net.cpp:84] Creating Layer derelu1_3d
I0310 00:29:06.824383  3796 net.cpp:406] derelu1_3d <- deconv1_3d
I0310 00:29:06.824388  3796 net.cpp:367] derelu1_3d -> deconv1_3d (in-place)
I0310 00:29:06.825677  3796 net.cpp:122] Setting up derelu1_3d
I0310 00:29:06.825692  3796 net.cpp:129] Top shape: 1 96 20 20 20 (768000)
I0310 00:29:06.825696  3796 net.cpp:137] Memory required for data: 22746412
I0310 00:29:06.825703  3796 layer_factory.hpp:77] Creating layer deconv0_3d
I0310 00:29:06.825716  3796 net.cpp:84] Creating Layer deconv0_3d
I0310 00:29:06.825721  3796 net.cpp:406] deconv0_3d <- deconv1_3d
I0310 00:29:06.825727  3796 net.cpp:380] deconv0_3d -> deconv0_3d
I0310 00:29:06.825764  3796 net.cpp:122] Setting up deconv0_3d
I0310 00:29:06.825769  3796 net.cpp:129] Top shape: 1 1 20 20 20 (8000)
I0310 00:29:06.825773  3796 net.cpp:137] Memory required for data: 22778412
I0310 00:29:06.825778  3796 layer_factory.hpp:77] Creating layer reconst
I0310 00:29:06.825784  3796 net.cpp:84] Creating Layer reconst
I0310 00:29:06.825788  3796 net.cpp:406] reconst <- deconv0_3d
I0310 00:29:06.825793  3796 net.cpp:380] reconst -> reconst
I0310 00:29:06.825799  3796 net.cpp:122] Setting up reconst
I0310 00:29:06.825803  3796 net.cpp:129] Top shape: 1 1 20 20 20 (8000)
I0310 00:29:06.825806  3796 net.cpp:137] Memory required for data: 22810412
I0310 00:29:06.825810  3796 net.cpp:200] reconst does not need backward computation.
I0310 00:29:06.825814  3796 net.cpp:200] deconv0_3d does not need backward computation.
I0310 00:29:06.825817  3796 net.cpp:200] derelu1_3d does not need backward computation.
I0310 00:29:06.825820  3796 net.cpp:200] deconv1_3d does not need backward computation.
I0310 00:29:06.825824  3796 net.cpp:200] derelu2_3d does not need backward computation.
I0310 00:29:06.825826  3796 net.cpp:200] deconv2_3d does not need backward computation.
I0310 00:29:06.825830  3796 net.cpp:200] derelu3_3d does not need backward computation.
I0310 00:29:06.825834  3796 net.cpp:200] deconv3_3d does not need backward computation.
I0310 00:29:06.825837  3796 net.cpp:200] derelu4_3d does not need backward computation.
I0310 00:29:06.825840  3796 net.cpp:200] deconv4_3d does not need backward computation.
I0310 00:29:06.825844  3796 net.cpp:200] reshape does not need backward computation.
I0310 00:29:06.825847  3796 net.cpp:200] small-2 does not need backward computation.
I0310 00:29:06.825851  3796 net.cpp:200] fc8-imfeat does not need backward computation.
I0310 00:29:06.825855  3796 net.cpp:200] drop7 does not need backward computation.
I0310 00:29:06.825858  3796 net.cpp:200] relu7 does not need backward computation.
I0310 00:29:06.825861  3796 net.cpp:200] fc7 does not need backward computation.
I0310 00:29:06.825865  3796 net.cpp:200] drop6 does not need backward computation.
I0310 00:29:06.825868  3796 net.cpp:200] relu6 does not need backward computation.
I0310 00:29:06.825871  3796 net.cpp:200] fc6 does not need backward computation.
I0310 00:29:06.825875  3796 net.cpp:200] pool5 does not need backward computation.
I0310 00:29:06.825878  3796 net.cpp:200] relu5 does not need backward computation.
I0310 00:29:06.825881  3796 net.cpp:200] conv5 does not need backward computation.
I0310 00:29:06.825886  3796 net.cpp:200] relu4 does not need backward computation.
I0310 00:29:06.825889  3796 net.cpp:200] conv4 does not need backward computation.
I0310 00:29:06.825893  3796 net.cpp:200] relu3 does not need backward computation.
I0310 00:29:06.825896  3796 net.cpp:200] conv3 does not need backward computation.
I0310 00:29:06.825899  3796 net.cpp:200] norm2 does not need backward computation.
I0310 00:29:06.825902  3796 net.cpp:200] pool2 does not need backward computation.
I0310 00:29:06.825906  3796 net.cpp:200] relu2 does not need backward computation.
I0310 00:29:06.825909  3796 net.cpp:200] conv2 does not need backward computation.
I0310 00:29:06.825912  3796 net.cpp:200] norm1 does not need backward computation.
I0310 00:29:06.825917  3796 net.cpp:200] pool1 does not need backward computation.
I0310 00:29:06.825919  3796 net.cpp:200] relu1 does not need backward computation.
I0310 00:29:06.825922  3796 net.cpp:200] conv1 does not need backward computation.
I0310 00:29:06.825927  3796 net.cpp:200] input does not need backward computation.
I0310 00:29:06.825929  3796 net.cpp:242] This network produces output reconst
I0310 00:29:06.825947  3796 net.cpp:255] Network initialization done.
I0310 00:29:14.732579  3796 net.cpp:744] Ignoring source layer data
I0310 00:29:14.732628  3796 net.cpp:744] Ignoring source layer label-voxel_data_1_split
I0310 00:29:14.732640  3796 net.cpp:744] Ignoring source layer scale256
I0310 00:29:14.732648  3796 net.cpp:744] Ignoring source layer conv1_3d
I0310 00:29:14.732657  3796 net.cpp:744] Ignoring source layer relu1_3d
I0310 00:29:14.732666  3796 net.cpp:744] Ignoring source layer conv2_3d
I0310 00:29:14.732673  3796 net.cpp:744] Ignoring source layer relu2_3d
I0310 00:29:14.732681  3796 net.cpp:744] Ignoring source layer conv3_3d
I0310 00:29:14.732688  3796 net.cpp:744] Ignoring source layer relu3_3d
I0310 00:29:14.732695  3796 net.cpp:744] Ignoring source layer conv4_3d
I0310 00:29:14.732702  3796 net.cpp:744] Ignoring source layer relu4_3d
I0310 00:29:14.732709  3796 net.cpp:744] Ignoring source layer small-1
I0310 00:29:14.789407  3796 net.cpp:744] Ignoring source layer loss
I0310 00:29:19.993075  3796 net.cpp:744] Ignoring source layer data
I0310 00:29:19.993105  3796 net.cpp:744] Ignoring source layer data_data_0_split
I0310 00:29:19.993109  3796 net.cpp:744] Ignoring source layer scale256
I0310 00:29:19.993113  3796 net.cpp:744] Ignoring source layer conv1_3d
I0310 00:29:19.993116  3796 net.cpp:744] Ignoring source layer relu1_3d
I0310 00:29:19.993120  3796 net.cpp:744] Ignoring source layer conv2_3d
I0310 00:29:19.993124  3796 net.cpp:744] Ignoring source layer relu2_3d
I0310 00:29:19.993126  3796 net.cpp:744] Ignoring source layer conv3_3d
I0310 00:29:19.993131  3796 net.cpp:744] Ignoring source layer relu3_3d
I0310 00:29:19.993135  3796 net.cpp:744] Ignoring source layer conv4_3d
I0310 00:29:19.993140  3796 net.cpp:744] Ignoring source layer relu4_3d
I0310 00:29:19.993147  3796 net.cpp:744] Ignoring source layer small-1
I0310 00:29:20.012852  3796 net.cpp:744] Ignoring source layer 
X Error of failed request:  BadValue (integer parameter out of range for operation)
  Major opcode of failed request:  154 (GLX)
  Minor opcode of failed request:  3 (X_GLXCreateContext)
  Value in failed request:  0x0
  Serial number of failed request:  49
  Current serial number in output stream:  50

How can I find the defination of “Average Precision”

Hello, thanks for your sharing and excellent work. In your paper, you tested your model over IKEA dataset and scored with "average precison" to represent the reconstruction ability of your trained model. And this critic is cited by several papers successively. However, there is no specific defination of Average Precision.
Could you provide the defination of "Average Precision" for me? Thank you!

error for num_axes() <=4

Hi,

I tried to run "python src/testing/reconst.py", and I got the follow error.

Check failed: num_axes() <= 4 (5 vs. 4) Cannot use legacy accessors on Blobs with > 4 axes

Note, I tried both the caffe version you listed and the latest version. But I still get the same error.

Could you please help me solve the issue?

Output CSV with mesh points instead of image

Hi,

Hope you are all well !

Was wondering if it could be possible to output a csv file with the object points of object detected as wanted to try it with unity and use procedural meshes to visualize them in 3D.

Would be awesome if you could help me or give a hint about how to do it with caffe as I have no "deep" experience in deep learning. :-)

Richard

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.