search

CSAILVision / sceneparsing

Development kit for MIT Scene Parsing Benchmark
http://sceneparsing.csail.mit.edu
BSD 3-Clause "New" or "Revised" License
461 stars 194 forks source link

Can't reproduce test results using pre-trained Dilated model #4

Closed ieted closed 8 years ago

ieted commented 8 years ago

Using the offered DilatedNet_iter_120000.caffemodel model and demoSegmentation.m script, I am unable to reproduce the qualitative test results posted at the bottom of README file (see here https://github.com/CSAILVision/sceneparsing#pre-trained-models-on-going). Here are two examples I got using DilatedNet caffe model.

Are the released models exactly the same with the ones you are using?

pengpaiSH commented 8 years ago

+1 I also found this problem! @hangzhaomit

xavierpuigf commented 8 years ago

There was an error in the demoSegmentation.m script. You should now be able to reproduce the segmentation results correctly.

Thanks for noticing

pengpaiSH commented 8 years ago

@xavierpuigf Thanks for your quick response. Would you mind having a look at the other issue that I asked?