Downloading data from imagenet

Question

I am told that the following list of "puppy" image URL's are from imagenet.

https://github.com/asharov/cute-animal-detector/blob/master/data/puppy-urls.txt

How do I download another category for e.g. "cats"?

Where can I get the entire list of imagenet categories along with their explanation in csv?

Necessitarianism

Thanks for suggesting to look into documentation. I managed to get the data that I was looking for. You can post it as an answer. – Necessitarianism

Answer 1

Unfortunately, ImageNet is no longer as easily accessible as it previously was. You now have to create a free account, and then request access to the database using an email address that demonstrates your status as a non-commercial researcher. Following is an excerpt of the announcement posted on March 11, 2021 (does not specifically address the requirements to obtain an account and request access permission but explains some of their reasons for changing the website generally).

We are proud to see ImageNet's wide adoption going beyond what was originally envisioned. However, the decade-old website was burdened by growing download requests. To serve the community better, we have redesigned the website and upgraded its hardware. The new website is simpler; we removed tangential or outdated functions to focus on the core use case—enabling users to download the data, including the full ImageNet dataset and the ImageNet Large Scale Visual Recognition Challenge (ILSVRC).

ORIGINAL ANSWER (LINKS NO LONGER VALID):

You can interactively explore available synsets (categories) in the /explore directory, each synset page has a "Downloads" tab where you can download category image URLs.

Alternatively, you can use the ImageNet API. You can download image URLs for a particular synset using the synset id or wnid. The image URL download link below uses the wnid n02121808 for domestic cat, house cat, Felis domesticus, Felis catus.

http://www.image-net.org/api/text/imagenet.synset.geturls?wnid=n02121808

You can find the wnid for a particular synset using the explore link above (the id for a selected synset will be displayed in the browser address bar).

You can retrieve a list of all available synsets (by id) from the no longer valid:

http://www.image-net.org/api/text/imagenet.synset.obtain_synset_list.

You can retrieve the words associated with any synset id as follows (another cat example).

http://www.image-net.org/api/text/wordnet.synset.getwords?wnid=n02121808

Emaciation

@Hugo - unfortunately, ImageNet has changed significantly since this answer was originally posted, see update. – Emaciation

this answer is no longer valid. some of the links in this post have been expired now – Pharos

@Pharos - see the first 2 paragraphs of the answer above the "ORIGINAL ANSWER (LINKS NO LONGER VALID)" heading. I removed the hyperlinks to make it more clear but left the link text to preserve the readability of the original answer. – Emaciation

Answer 2

or you can download smaller size of imagenet, mini-imagenet:

1. https://github.com/yaoyao-liu/mini-imagenet-tools

2-1. https://github.com/dragen1860/LearningToCompare-Pytorch/issues/4

2-2. https://github.com/twitter/meta-learning-lstm/tree/master/data/miniImagenet

Ellary

Answer 3

You can easily use the python package MLclf to download and transform the mini-imagenet data for the traditional image classification task or the meta-learning task. just use:

pip install MLclf

You can also see for more details:

https://pypi.org/project/MLclf/

Miche

While this link may answer the question, it is better to include the essential parts of the answer here and provide the link for reference. Link-only answers can become invalid if the linked page changes. - From Review – Prerequisite