robert0812 / deepsaldet Goto Github PK

Hi,
I cannot download the models via Dropbox. It seems that Dropbox has temporarily disabled the account as it is generating too much traffic.
Is there other way rather than Baidu or Dropbox to share the models?
Thank you,

I can't get model files.
www.ee.cuhk.edu.hk/~rzhao/sharedata/model_imagenet_bbox_pretrain_iter_780000_cascade_stage2_iter_80000.caffemodel
www.ee.cuhk.edu.hk/~rzhao/sharedata/model_imagenet_padimage_pretrain_iter_600000_stage1_iter_80000.caffemodel

thanks.

Hi, in figure 3: the output of global-context branch is fc8 with 2 neurons, but how can it be a binary map and downsampled to 64x64 and then flattened into 4096 vector? Also, as mentioned in the paper

Superpixel segmentation is firstly performed on images using the SLIC method, and the input of global-context CNN is a superpixel-centered large context window including the full image.

Is there only one superpixel per image? I found SLIC would generate many superpixels for each image, Can u tell more detail on this part? thanks.

could you offer the pytorch code?
thanks

Hi, I can successfully compile the caffe from caffe.berkeleyvision.org/. But I failed to compile the caffe in caffe-sal folder and got the following problem:

/bin/ld: cannot find -lblas
collect2: error: ld returned 1 exit status
make: *** [.build_release/lib/libcaffe.so] Error 1

Is anything important that i missed?

Hi,

I am facing errors while compiling the caffe contained in the caffe-sal folder.
running the command 'make all' results in error.
How do I compile the caffe-sal version of caffe. Any help would be really appreciated. Thanks.

The caffe I compiled is the caffe-rc5 given by the official website. Can you run your code? If you need to make changes, where do you want to make changes, and need to recompile?

Got this error running on a CPU machine (which had Caffe installed earlier), after following all the steps listed -

F1015 23:31:27.377449  4597 common.cpp:53] CPU-only Mode
*** Check failure stack trace: ***
    @     0x7efe9b0e9dcd  google::LogMessage::Fail()
    @     0x7efe9b0ebfb2  google::LogMessage::SendToLog()
    @     0x7efe9b0e9953  google::LogMessage::Flush()
    @     0x7efe9b0ec8ad  google::LogMessageFatal::~LogMessageFatal()
    @           0x445960  caffe::Caffe::SetDevice()
    @           0x4108b4  feature_extraction_pipeline<>()
    @     0x7efe98661ec5  __libc_start_main
    @           0x40bdd7  (unknown)
Aborted (core dumped)
Traceback (most recent call last):
  File "predict_stage1.py", line 161, in <module>
    main()
  File "predict_stage1.py", line 138, in main
    outputs = leveldb2ndarray(dbfolder=leveldb_folder, num_feat=num_window, dim_feat=2)
  File "predict_stage1.py", line 42, in leveldb2ndarray
    dt.ParseFromString(db.Get('%d' %(idx)))
KeyError
Exception TypeError: TypeError("'NoneType' object is not callable",) in <bound method UmfpackContext.new_del of <scipy.sparse.linalg.dsolve.umfpack.umfpack.UmfpackContext object at 0x7feaa8df2350>> ignored
cp: cannot stat ‘_tmp/smaps_sc/*’: No such file or directory

Any ideas what might be going wrong?

Loading images ...
Superpixel segmentation ...
Extracting windows from 0-th image ... [Total: 10]
Extracting windows from 1-th image ... [Total: 10]
Extracting windows from 2-th image ... [Total: 10]
Extracting windows from 3-th image ... [Total: 10]
Extracting windows from 4-th image ... [Total: 10]
Extracting windows from 5-th image ... [Total: 10]
Extracting windows from 6-th image ... [Total: 10]
Extracting windows from 7-th image ... [Total: 10]
Extracting windows from 8-th image ... [Total: 10]
Extracting windows from 9-th image ... [Total: 10]
1023
E0702 16:22:40.293716 7705 extract_features.cpp:49] Using GPU
E0702 16:22:40.293944 7705 extract_features.cpp:55] Using Device_id=0
E0702 16:22:42.401402 7705 extract_features.cpp:135] Extacting features for 1024 images ...
E0702 16:22:45.945432 7705 extract_features.cpp:173] Extracted features of 1000 query images for feature blob fc8_msra10k
E0702 16:22:46.005468 7705 extract_features.cpp:185] Extracted features of 1024 query images for feature blob fc8_msra10k
E0702 16:22:46.005511 7705 extract_features.cpp:189] Successfully extracted the features!
Exception TypeError: TypeError("'NoneType' object is not callable",) in <bound method UmfpackContext.new_del of <scipy.sparse.linalg.dsolve.umfpack.umfpack.UmfpackContext object at 0x7f9c0aa0aa50>> ignored

The saliency images are generated, but with this error information. Is it OK?

After 'loading images' and 'superpixel segmentation', the process will be killed, and the error is 'can't get the stat of '_tmp/smaps_sc/*: no that file or directory''..
Did i miss something important?