search

valeoai / 3DGenZ

Public repository of the 3DV 2021 paper "Generative Zero-Shot Learning for Semantic Segmentation of 3D Point Clouds"
Other
34 stars 3 forks source link

Question about the generator GMMN #3

Closed Daisy-GENG closed 2 years ago

Daisy-GENG commented 2 years ago

Thank you for publishing your code of such an interesting work. I am new to ZSL/G-ZSL, so I am little bit confused about the source of word representations(Word2Vec/GloVe) of the SemanticKITTI dataset that you provided in this repo. Do you manually build up the corpus of SemanticKITTI dataset, and use Word2Vec/GloVe to get these word representations, or is it publicly available?

Thank you so much!!

Best regards, RUOYU GENG

BjoernMichele commented 2 years ago

Thank you very much for your interest in our work.

The word representations for the semantic segmentation datasets are obtained from publicly available word2vec and GloVe models. So we querried the pre-trained word2vec/GloVe models with the class names to obtain the vector representations for our classes. I have to double check but for W2V I am pretty sure we used the model from the gensim library (https://radimrehurek.com/gensim/models/word2vec.html), and for GloVe the models provided on the official website (https://nlp.stanford.edu/projects/glove/).

Best regards, Bjoern

Daisy-GENG commented 2 years ago

Dear Bjoern,

Thank you for your reply!

Best regards, RUOYU