pedropro / TACO

🌮 Trash Annotations in Context Dataset Toolkit

Home Page:http://tacodataset.org

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Good peoject!

JTDeve opened this issue · comments

This is an interesting project,but dataset is not enough,I want to join you!

Thanks! You are right. We are about to: (1) add at least more 3000 images for annotation, (2) open our annotation service to public: http://tacodataset.org/annotate and (3) add more code for data augmentation and object transplanting.
This should happen this month. I will keep posting updates on twitter.

Our first goal is to hit 10000 annotated images. I estimate this would be enough to get decent detection results with transfer learning for some classes, e.g, cans, bottles, cups, straws, plastic lids.

The best way to help right now would be to start annotating (once this is open). Secondly, collect and share litter images, e.g. via flickr. We will acknowledge fruitful collaborations and whoever produces independently any significant work, e.g., tools, annotations, results, models, applications.

Hi Pedro!
I'm very glad that I'd found your project. It's really helpful for data scientist. Special thanks to you and your teamates.
I'm working on a small project that need part of your image dataset. Can you show me how to download the images? It's been two days and I still couldn't find the way to download it. I've tried to type "python3 download.py" in cmd window at the directory of this downloaded project but i saw nothing.
I'm really appreciate your work and looking forward to see your response.
Thank you so much for speding your time reading this and repy.

Hey Khoa! Can you open a new issue "Download dataset on windows" and post more details, including your python version and terminal output. I am going to try to replicate this on Windows.

Hey Khoa! Can you open a new issue "Download dataset on windows" and post more details, including your python version and terminal output. I am going to try to replicate this on Windows.

I'm on it!