Skip to content
A Neural Compositional Paradigm for Image Captioning
Lua Python Java Shell
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
coco-caption
data
image_models
misc
nlp
scripts
tests
.gitignore
LICENSE
README.md
compcap.png
encode_nounphrase.py
feat_eval_bibeamsearch.lua
feat_eval_nounphrase.lua
feat_train_connectingmodule.lua
feat_train_nounphrase.lua
feat_train_stoper.lua
samples.png

README.md

Code of A Neural Compositional Paradigm for Image Captioning

Borrowed some code from AdaptiveAttention.

Overview

Overview

Summarization

  • We propose a compositional captioning model, where caption generation follows a recursive structure, which naturally fits the properties of human language. Specifically, we could generate diverse captions based on the same visual semantics by changing syntax, or attentions.

Good

Codes

  • nlp: scripts for processing sentences using a nlp toolkit

  • data: scripts for processing data

  • misc: models, dataloaders, and auxiliary codes

  • image_models: pretrained image models

  • test: scripts for studies

  • scripts: bashs

Citation

@inproceedings{dai2018aneural,
title = {A Neural Compositional Paradigm for Image Captioning},
author = {Dai, Bo and Fidler, Sanja and Lin, Dahua},
booktitle = {Advances in Neural Information Processing Systems 31},
pages = {658--668},
year = {2018}
}
You can’t perform that action at this time.