search

braindeadpool / bf-vos

Pytorch implementation of Blazingly Fast Video Object Segmentation with Pixel-Wise Metric Learning (Chen et al)
27 stars 6 forks source link

Data Preparation #3

Open haofengac opened 6 years ago

haofengac commented 6 years ago

Hi,

I'm confused about the structure of the DAVIS folder. When I run the code it keeps telling me there's no train.txt. Where should I find it?

Thanks

braindeadpool commented 6 years ago

It should be at bfvos/dataset/DAVIS/<datasettype>/ImageSets/train.txt where datasettype is one of trainval, testdev or testchallenge.

emedinac commented 5 years ago

I downloaded the dataset from the official website and I have some problems. I don't have the < datasettype > folder. I only see the following folders: Annotations, Annotations_semantics, ImageSets, JPEGImages

am I doing something wrong in the data organization when I unzip the files?

emedinac commented 5 years ago

Sorry for the trouble... I solved the problem. I used this link to download the dataset "https://davischallenge.org/davis2016/code.html" which is incorrect for this repository. The correct link to download is this "https://davischallenge.org/davis2017/code.html". Inside this, 2016 and 2017 datasets are available.