search

kaixin96 / PANet

Code for our ICCV 2019 paper PANet: Few-Shot Image Semantic Segmentation with Prototype Alignment
315 stars 64 forks source link

'SBD' folder -sbd_instance_process.py #28

Closed phuongchi911 closed 4 years ago

phuongchi911 commented 4 years ago

Hi Kaixin sorry to trouble u. May i ask what is your 'SBD' folder and how i can get it? Sorry if i missed out something. From my undertanding it seemed u already have the SBD folder in place and process them and save them toSegmentationClassAug folder. Thank you very much.

sbd_instance_process.py

and transform it from .mat to .png. Then transformed
images will be saved in VOC data folder. The name of
the new folder is "SegmentationObjectAug"
kaixin96 commented 4 years ago

Hi @phuongchi911 , the segmentation masks in SegmentationClassAug folder is downloaded from here. I only process SBD data for generating instance-level masks. You may download the dataset from here and use the folder benchmark_RELEASE/dataset as SBD folder.

Thank you.

phuongchi911 commented 4 years ago

Hi @phuongchi911 , the segmentation masks in SegmentationClassAug folder is downloaded from here. I only process SBD data for generating instance-level masks. You may download the dataset from here and use the folder benchmark_RELEASE/dataset as SBD folder.

Thank you.

Thanks so much for the details. Yes i have downloaded your SegmentationClassAug. May I ask if my data only has the masks photos like in the folder SegmentationClass, is it required to transform them to the SBD data? Thanks so much for the advise.

kaixin96 commented 4 years ago

Basically the data in the Google Drive is all you need to run the experiments.

sbd_instance_process.py is only needed when you already have SBD data and don't want to download large files from Google Drive. Then you can use it to transform the SBD data format (.mat) into VOC format (PNG image with mode 'P'). The code in this repo uses VOC format masks.

Regarding your questions, it is not needed to transform them to the SBD data.

Thank you.

kaixin96 commented 4 years ago

I’m closing this issue because it has been inactive for a while. Feel free to reopen if you have questions. Thank you.