lichengunc / refer Goto Github PK
View Code? Open in Web Editor NEWReferring Expression Datasets API
License: Apache License 2.0
Referring Expression Datasets API
License: Apache License 2.0
Hi,
The given segmentation data is a list of numbers. How can I visualize them?
Thanks
Why am I not getting the corresponding data set after running the refer.py file?
Are the lists of attributes for the analysis in paragraph 4.3 available anywhere? And your template-based parser?
Hi Li Cheng, do you keep a leaderboard of the various attempts at these datasets? I'm trying to figure out which one is the SOA right now. Really appreciate any pointers.
Is the training code and pre-trained model of your ECCV 2016 paper available to release?
Thanks,
Hi Licheng, thank you for your work.
It seems like all download links are not working since a few days ago, could you check and update the links? Thank you!
such as:
https://bvisionweb1.cs.unc.edu/licheng/referit/data/refclef.zip
https://bvisionweb1.cs.unc.edu/licheng/referit/data/refcoco.zip
https://bvisionweb1.cs.unc.edu/licheng/referit/data/refcoco+.zip
https://bvisionweb1.cs.unc.edu/licheng/referit/data/refcocog.zip
Hi, thanks for sharing your API. It's very easy to use.
However, this download link (https://bvisionweb1.cs.unc.edu/licheng/referit/data/images/saiapr_tc-12.zip) for refclef images are down.
Could you provide another link?
Thanks in advance.
I would like to have a script for how to generate refclef.zip.
Error:
Error compiling Cython file: ------------------------------------------------------------ ... cdef np.ndarray[np.double_t, ndim=1] np_poly n = len(poly) Rs = RLEs(n) for i, p in enumerate(poly): np_poly = np.array(p, dtype=np.double, order='F') rleFrPoly( <RLE*>&Rs._R[i], <const double*> np_poly.data, len(np_poly)/2, h, w ) ^ ------------------------------------------------------------ external/_mask.pyx:258:78: Cannot assign type 'double' to 'siz' Traceback (most recent call last): File "setup.py", line 24, in <module> ext_modules=cythonize(ext_modules) File "/opt/miniconda3/envs/pytorch1.6/lib/python3.6/site-packages/Cython-3.0a6-py3.6-linux-x86_64.egg/Cython/Build/Dependencies.py", line 1110, in cythonize cythonize_one(*args) File "/opt/miniconda3/envs/pytorch1.6/lib/python3.6/site-packages/Cython-3.0a6-py3.6-linux-x86_64.egg/Cython/Build/Dependencies.py", line 1277, in cythonize_one raise CompileError(None, pyx_file) Cython.Compiler.Errors.CompileError: external/_mask.pyx
python version =3.6, cython = 3.0a6
When visiting the web "https://bvisionweb1.cs.unc.edu/licheng/referit/data/images/saiapr_tc-12.zip", it turn out to be like this:
How to fix it?
This repo doesn't have a LICENSE file. Please add one so others can use the code and contribute back, like the apache license.
I was just wondering, instead of copying the code inside ./external
. Why not just use pycocotools from pip?
The readme says "add "mscoco" into the data/images folder, which can be from mscoco"
The readme should also specify the version of the dataset since there are multiple. I assume it's train2014 because of the images/mscoco/images/train2014 line in refer.py.
Hello, I found annotations extracted by REFER class in refer.py seems to have some problems regarding categories. When I try to check whether annotations on category are correct, I found the inconsistency between the category dictionary and referring query. For example, the query (target object) is about sky, while its corresponding category is chimney. Would you please check annotations, mainly in 'refclef/instances.json'? Hoping you could reply soon :)
Is refcocog and Grefcoco dataset same?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.