Following the readme steps, running eval_box.py also raises an exception.

python eval_box.py test.txt ../models/lpo_VOC_0.1.dat
Traceback (most recent call last):
  File "eval_box.py", line 57, in <module>
    bo,pool_s = evaluateBox( prop, over_segs, boxes, name='(tst)' )
  File "eval_box.py", line 43, in evaluateBox
    print( "LPO %05s & %d & %0.3f & %0.3f & %0.3f & %0.3f & %0.3f \\\\"%(name,np.nanmean(pool_s),np.mean(bo),np.mean(bo>=0.5), np.mean(bo>=0.7), np.mean(bo>=0.9), np.mean(2*np.maximum(bo-0.5,0)) ) )
UnboundLocalError: local variable 'pool_s' referenced before assignment

both bo, pool_s are dangerously non-initialized.

I guess something else is off, but I cannot quite figure out what, since none of the scripts seems to work as expected.

I followed the install pipeline and no error. But at last when I run the command-''bash eval_all.sh', the console printed "ImportError: No module named pylab". Then I changed the eval_all.sh script file, all python3 to python. Though there wasn't previous error, the console printed a new error, "ImportError: No module named python.lpo". Now I had no idea and hope you could help me. Thanks.

I followed the instructions in the readme and compiled with -DUSE_PYTHON=2 , when I do bash eval_all.sh I get error:

Traceback (most recent call last):
File "train_lpo.py", line 31, in
from lpo import *
File "/home/revathy/lpo-release/src/lpo.py", line 45, in
from python.lpo import *
ImportError: dynamic module does not define init function (PyInit_lpo)

hope you could help me. Thanks.

Hi, @philkr
I think there is a typos in matlab/CMakeLists.txt

add_library( gop_mex SHARED gop_mex.cpp )
target_link_libraries( gop_mex util imgproc learning contour segmentation proposals gomp )

should be

add_library( lpo_mex SHARED lpo_mex )
target_link_libraries( lpo_mex util imgproc learning contour segmentation proposals gomp )

As you do not have gop_mex.cpp in this project, Do you?

When I execute ./sam_eval_all.sh which just change to use python2.7 , it shows:

sam@sam-desktop:~/code/download/Segmentation/lpo/src$ ./sam_eval_all.sh
File '/path/to/datasets/VOC2012/ImageSets/Segmentation/val.txt' not found! Check if DATA_DIR is set properly.
Traceback (most recent call last):
File "train_lpo.py", line 139, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 70, in loadAndOverSegDataset
data = loader()
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
ValueError: Failed to load dataset
File '/path/to/datasets/VOC2012/ImageSets/Segmentation/val.txt' not found! Check if DATA_DIR is set properly.
Traceback (most recent call last):
File "train_lpo.py", line 139, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 70, in loadAndOverSegDataset
data = loader()
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
ValueError: Failed to load dataset
File '/path/to/datasets/VOC2012/ImageSets/Segmentation/val.txt' not found! Check if DATA_DIR is set properly.
Traceback (most recent call last):
File "train_lpo.py", line 139, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 70, in loadAndOverSegDataset
data = loader()
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
ValueError: Failed to load dataset
File '/path/to/datasets/VOC2012/ImageSets/Segmentation/val.txt' not found! Check if DATA_DIR is set properly.
Traceback (most recent call last):
File "train_lpo.py", line 139, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 70, in loadAndOverSegDataset
data = loader()
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
ValueError: Failed to load dataset
File '/path/to/datasets/VOC2012/ImageSets/Segmentation/val.txt' not found! Check if DATA_DIR is set properly.
Traceback (most recent call last):
File "train_lpo.py", line 139, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 70, in loadAndOverSegDataset
data = loader()
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
ValueError: Failed to load dataset
File '/path/to/datasets/VOC2012/ImageSets/Segmentation/val.txt' not found! Check if DATA_DIR is set properly.
Traceback (most recent call last):
File "train_lpo.py", line 139, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 70, in loadAndOverSegDataset
data = loader()
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 98, in
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )

ValueError: Failed to load dataset

sam@sam-desktop:~/code/download/Segmentation/lpo/src$ cat ./sam_eval_all.sh

This script reproduces table 3 in the paper

python train_lpo.py -f0 0.2 ../models/lpo_VOC_0.2.dat
python train_lpo.py -f0 0.1 ../models/lpo_VOC_0.1.dat
python train_lpo.py -f0 0.05 ../models/lpo_VOC_0.05.dat
python train_lpo.py -f0 0.03 ../models/lpo_VOC_0.03.dat
python train_lpo.py -f0 0.02 ../models/lpo_VOC_0.02.dat
python train_lpo.py -f0 0.01 ../models/lpo_VOC_0.01.dat -iou 0.925 # Increase the IoU a bit to make sure the number of proposals match
sam@sam-desktop:~/code/download/Segmentation/lpo/src$

How to solve it? Thank you~

It's great that you made easy-to-use code to evaluate on multiple datasets. However, I'm having a little trouble evaluating on individual images for my own application.

Could you make a simple python example showing how to use this library for evaluating on an individual image?

Thanks!

Hi,

Does this library require GOP library? because the cmakelists.txt of matlab folder contains:

Hi! The paper mentions that models are trained for both Pascal VOC and COCO, as the COCO data tends to be a small part of the image, and a different model is needed to segment such small parts. In the data folder, there are files that have VOC in the name. Are the COCO models in this library already too, or do we need to train on COCO to generate those?

On 10.10.3, with Python 2.7.10 and boost and boost-python 1.58.0. Built using:

cmake .. -DCMAKE_BUILD_TYPE=Release -DDATA_DIR=~/test_images/coco-master/images/val2014 -DUSE_PYTHON=2
make -j9

Saw some warnings only, such as:
In file included from /Users/peterwang/CPP_Resources/lpo-release/lib/crf/crf.cpp:31:
/Users/peterwang/CPP_Resources/lpo-release/external/ibfs/ibfs.h:161:2: warning: 'Node' defined as a class here but previously declared as a struct
[-Wmismatched-tags]
class Node
^
/Users/peterwang/CPP_Resources/lpo-release/external/ibfs/ibfs.h:150:2: note: did you mean class here?
struct Node;
^~~~~~
class

Tried:

python train_lpo.py -f0 0.2 ../models/lpo_VOC_0.2.dat
and got: Segmentation fault: 11
This appears to have crashed on "from python.lpo import *" in lpo.py.

Crash report contained:
...
Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0 ??? 000000000000000000 0 + 0
1 org.python.python 0x0000000103d150dd PyEval_GetGlobals + 23
2 org.python.python 0x0000000103d2462b PyImport_Import + 137
3 org.python.python 0x0000000103d22d27 PyImport_ImportModule + 31
4 lpo.so 0x00000001033f45a3 init_numpy() + 19
5 lpo.so 0x00000001033f4779 defineUtil() + 25
6 lpo.so 0x00000001033f4499 init_module_lpo() + 9
7 libboost_python-mt.dylib 0x0000000103c36391 boost::python::handle_exception_impl(boost::function0) + 81
8 libboost_python-mt.dylib 0x0000000103c373b9 boost::python::detail::init_module(char const_, void (_)()) + 121
9 org.python.python 0x0000000101836327 _PyImport_LoadDynamicModule + 140
...

Saw the note in external/boost/readme.txt:
"In order to use a non-system boost library copy the "boost" and "libs" directory of a recent boost release (eg 1.57) here."

And in build/lib/python/CMakeFiles/lpo.dir/depend.make:
...
lib/python/CMakeFiles/lpo.dir/boost.cpp.o: /usr/local/include/boost/array.hpp
lib/python/CMakeFiles/lpo.dir/boost.cpp.o: /usr/local/include/boost/assert.hpp
lib/python/CMakeFiles/lpo.dir/boost.cpp.o: /usr/local/include/boost/bind.hpp
...

These seem to suggest the seg fault was due to boost version mismatch?

Is it sufficient to just do:

ln -s /usr/local/Cellar/boost/1.58.0 external/boost/
ln -s /usr/local/Cellar/boost/1.58.0/lib external/boost/libs

Or something else?

BTW, boost and boost-python were installed as part of setting up Caffe. The Caffe ImageNet model ran successfully when invoked from a Python test app.

Thanks for any light you could help shed.

[ 26%] Building CXX object lib/util/CMakeFiles/util.dir/geodesics.cpp.o
In file included from /media/dat1/liao/lpo/lib/util/threading.cpp:27:0:
/media/dat1/liao/lpo/lib/util/threading.h: In member function ‘void ThreadedQueue<T>::process(ThreadedQueue<T>::F, const std::vector<T>&)’:
/media/dat1/liao/lpo/lib/util/threading.h:188:5: error: ‘sleep_for’ is not a member of ‘std::this_thread’
make[2]: *** [lib/util/CMakeFiles/util.dir/threading.cpp.o] Error 1
make[2]: *** Waiting for unfinished jobs....
In file included from /media/dat1/liao/lpo/lib/util/algorithm.h:32:0,
from /media/dat1/liao/lpo/lib/util/algorithm.cpp:27:
/media/dat1/liao/lpo/lib/util/threading.h: In member function ‘void ThreadedQueue<T>::process(ThreadedQueue<T>::F, const std::vector<_RealType>&)’:
/media/dat1/liao/lpo/lib/util/threading.h:188:5: error: ‘sleep_for’ is not a member of ‘std::this_thread’
make[2]: *** [lib/util/CMakeFiles/util.dir/algorithm.cpp.o] Error 1
make[1]: *** [lib/util/CMakeFiles/util.dir/all] Error 2
make: *** [all] Error 2

When I tried to train LPO on the COCO dataset using the command:
python train_lpo.py ../models/lpo_COCO_0.02.dat -b -t -f0 0.02
I get the following error:
Traceback (most recent call last):
File "train_lpo.py", line 121, in
boxes = [proposals.Proposals(s,np.eye(np.max(s)+1).astype(bool)).toBoxes() for s in segmentations]
File "/home/anaconda/lib/python2.7/site-packages/numpy/core/fromnumeric.py", line 2135, in amax
out=out, keepdims=keepdims)
File "/home/anaconda/lib/python2.7/site-packages/numpy/core/_methods.py", line 26, in _amax
return umr_maximum(a, axis, None, out, keepdims)
ValueError: operands could not be broadcast together with shapes (10,2) (5,2)

Without the -b argument, I was able to successfully train LPO on the COCO dataset. I'm using Mac OS X 10.10.3.

Hello. I am very excited of your previous approach: Geodesic Object Proposals. And there you propose good example of using library in C++ code. Can you give some part of code for use this library? I found that they very similar.

Thank you.

When running analyze_model.py (with python 2.7) over the demonstratation model, I get a division by zero.

python analyze_model.py ../models/lpo_VOC_0.1.dat 
/home/rodrigob/.local/lib/python2.7/site-packages/numpy/core/_methods.py:55: RuntimeWarning: Mean of empty slice.
  warnings.warn("Mean of empty slice.", RuntimeWarning)
/home/rodrigob/.local/lib/python2.7/site-packages/numpy/core/_methods.py:67: RuntimeWarning: invalid value encountered in double_scalars
  ret = ret.dtype.type(ret / rcount)
Traceback (most recent call last):
  File "analyze_model.py", line 115, in <module>
    evaluateDetailed( prop, over_segs, segmentations )
  File "analyze_model.py", line 101, in evaluateDetailed
    print( names[m], '&', np.mean(ps[m]), '&', np.mean(bo[m]>=bbo)*100, '&', np.sqrt(np.mean(ma[m])), '&', t[m]/len(ma[m]) )
ZeroDivisionError: integer division or modulo by zero

I am not getting this right from the help and readme ?

When I was building the program, I got an error like :

**collect2: error: ld returned 1 exit status

examples/CMakeFiles/example.dir/build.make:97: recipe for target 'examples/example' failed
make[2]: *** [examples/example] Error 1
CMakeFiles/Makefile2:729: recipe for target 'examples/CMakeFiles/example.dir/all' failed
make[1]: *** [examples/CMakeFiles/example.dir/all] Error 2
[ 97%] Built target gop
Makefile:116: recipe for target 'all' failed
make: * [all] Error 2

How to resolve my problem?
Thank you very much!

It seems gcc 4.8 is required to support the vectorized array subscripts. With gcc 4.7 I'm getting:

lpo/lib/imgproc/color.cpp:77:31: error: invalid types ‘__m128 {aka __vector(4) float}[int]’ for array subscript

Should the README be updated, or am I doing anything wrong ?

On win8 with 64 bit, I run compile.m, and modify the line
< 'cmd = ['mex -DLBFGS_FLOAT=32 -DEIGEN_DONT_PARALLELIZE -DNO_IMREAD -I../lib -I',EIGEN_DIR,' -I../external/liblbfgs-1.10/include/ lpo_mex.cpp ',' ../external/liblbfgs-1.10/lib/lbfgs.c ', all_files]' >.
But it fails to build the lpo_mex.cpp. I try it, but some problems are present. Can you find this bug?

Hello, thanks for sharing codes!
I compile your code success with python 2.7.
I also change python3 to python in eval_all.sh file.
When I run eval_all.sh, it shows:

Traceback (most recent call last):
File "train_lpo.py", line 129, in
prop.load(save_name)
ValueError: Unknown model type 'GlobalCRFModel'!
Traceback (most recent call last):
File "train_lpo.py", line 129, in
prop.load(save_name)
ValueError: Unknown model type 'GlobalCRFModel'!
Traceback (most recent call last):
File "train_lpo.py", line 129, in
prop.load(save_name)
ValueError: Unknown model type 'GlobalCRFModel'!
Traceback (most recent call last):
File "train_lpo.py", line 129, in
prop.load(save_name)
ValueError: Unknown model type 'GlobalCRFModel'!
Traceback (most recent call last):
File "train_lpo.py", line 129, in
prop.load(save_name)
ValueError: Unknown model type 'GlobalCRFModel'!
Traceback (most recent call last):
File "train_lpo.py", line 129, in
prop.load(save_name)
ValueError: Unknown model type 'GlobalCRFModel'!
sam@sam-desktop:~/code/download/Segmentation/lpo/src$

How to solve it?
Thank you very much~

Hey,

I've been trying to get a small bounding-box example working, based on your propose_hf5.py code.

Here is my code:

imgs = [ lpo.imgproc.imread('cat.jpg') ]

prop = lpo.proposals.LPO()
prop.load( 'dats/lpo_VOC_0.02.dat' )

detector = lpo.contour.MultiScaleStructuredForest()
detector.load( 'dats/sf.dat' )

over_segs = lpo.segmentation.generateGeodesicKMeans( detector, imgs, 1000 )

props = prop.propose( over_segs, 0.01, True )

props = props[0][0]

fig = plt.figure()
ax = fig.add_subplot(1, 1, 1)
ax.imshow(imgs[0])
for bb in props.toBoxes():
    ax.add_patch(matplotlib.patches.Rectangle((bb[0],bb[1]),bb[2],bb[3], color='red', fill=False))

I end up with:

If I play around with some of the parameters, I end up getting an enormous amount of proposal boxes.

I was hoping somebody could provide some advice to help me get this working.

Hello,
I run sed -e 's:python3:python:g' eval_all.sh > sam_eval_all.sh
And here is the output when execute sam_eval_all.sh:

Traceback (most recent call last):
File "train_lpo.py", line 137, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 94, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 52, in loadAndOverSegDataset
except FileNotFoundError:
NameError: global name 'FileNotFoundError' is not defined
Traceback (most recent call last):
File "train_lpo.py", line 137, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 94, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 52, in loadAndOverSegDataset
except FileNotFoundError:
NameError: global name 'FileNotFoundError' is not defined
Traceback (most recent call last):
File "train_lpo.py", line 137, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 94, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 52, in loadAndOverSegDataset
except FileNotFoundError:
NameError: global name 'FileNotFoundError' is not defined
Traceback (most recent call last):
File "train_lpo.py", line 137, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 94, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 52, in loadAndOverSegDataset
except FileNotFoundError:
NameError: global name 'FileNotFoundError' is not defined
Traceback (most recent call last):
File "train_lpo.py", line 137, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 94, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 52, in loadAndOverSegDataset
except FileNotFoundError:
NameError: global name 'FileNotFoundError' is not defined
Traceback (most recent call last):
File "train_lpo.py", line 137, in
over_segs,segmentations,boxes,names = loadVOCAndOverSeg( 'test', detector='mssf' )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 94, in loadVOCAndOverSeg
return loadAndOverSegDataset( lambda: ldr(im_set=="train",im_set=="valid",im_set=="test"), "VOC%s_%s"%(year,im_set), detector=detector, N_SPIX=N_SPIX )
File "/home/sam/code/download/Segmentation/lpo/src/util.py", line 52, in loadAndOverSegDataset
except FileNotFoundError:
NameError: global name 'FileNotFoundError' is not defined

Thank you~

Can I use this on a custom dataset that I have? Thanks!

Hey, whenever I try to compile using cmake it throws an error. Below is the output. Any help is appreciated.

CMake Error at /usr/local/share/cmake-3.2/Modules/FindPackageHandleStandardArgs.cmake:138 (message):
Could NOT find PythonInterp: Found unsuitable version "1.4", but required
is at least "3.3" (found /usr/bin/python3)
Call Stack (most recent call first):
/usr/local/share/cmake-3.2/Modules/FindPackageHandleStandardArgs.cmake:372 (_FPHSA_FAILURE_MESSAGE)
/usr/local/share/cmake-3.2/Modules/FindPythonInterp.cmake:162 (FIND_PACKAGE_HANDLE_STANDARD_ARGS)

lib/python/CMakeLists.txt:10 (find_package)

...I'm running Python 3.4 and CMake 3.2.3.

@philkr

i got the code installed and could get it working - I rain eval_all.sh and it seemed to run fine.

but i seem to be a bit lost post that. Specifically, I am looking for some simple steps on how to get this working for a custom data set (say even VOC 2007), step-by-step.

To get the box proposals using our own datasets, I think we need to have the equivalent sf.dat and the lpo_VOC.xx.dat files as inputs. But, how to generate these 2 for our own datasets?

When I ran train_lpo.py -f0 0.05 -t, it only created a VOC_2007_train_mssf_1000.dat in the /tmp folder. And I couldn't load it as I got an "out of memory error"

ImportError: /home/ksivakumar/lpo/build/lib/python/lpo.so: undefined symbol: png_set_longjmp_fn

...I am getting this error and I'm using libpng 1.5, which does have support for png_set_longjpm_fn. Any ideas?

Hi @philkr, and other lpo users,

Is there any way, convenient or not, to convert an ndarray object to the Image8u type used in this library? Using the lpo.improc.imread function obviously reads a file from disk, and provides the proper data type. I'm wondering if I can avoid having to read from disk if it's already in memory as an ndarray?

Thanks