SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1609.04938 · 2016

Image-to-Markup Generation with Coarse-to-Fine Attention

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 13 functions out of this paper's own repositories and ran 1 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
harvardnlp/im2markup canonical 0 of 1
da03/Attention-OCR reimplementation 1 of 1
jvpoulos/Attention-OCR pwc_unofficial 0 of 8
CT2001/image-to-latex pwc_unofficial 0 of 3
FunctionStatusWhere it lives
process_args Ran da03/Attention-OCR/src/launcher.py
code served (permissive licence) · get_code("5f977334a9330cac")
FMM_func Not yet run CT2001/image-to-latex/data-preprocess/no_chinese.py
code served (permissive licence) · get_code("5bfcddf02c8fcf58")
find_first Not yet run CT2001/image-to-latex/image_to_latex/models/resnet_transformer.py
code served (permissive licence) · get_code("d3fb41b62eaedad7")
generate_square_subsequent_mask Not yet run CT2001/image-to-latex/image_to_latex/models/resnet_transformer.py
code served (permissive licence) · get_code("9fae5025ebafa98e")
get_wmt_enfr_dev_set Not yet run jvpoulos/Attention-OCR/src/model/data_utils.py
code served (permissive licence) · get_code("ec532e8f70289f38")
get_wmt_enfr_train_set Not yet run jvpoulos/Attention-OCR/src/model/data_utils.py
code served (permissive licence) · get_code("0a7b6ca2e2bf4659")
max_2x1pool Not yet run jvpoulos/Attention-OCR/src/model/cnn.py
code served (permissive licence) · get_code("3ca4a31edc44c233")
max_2x2pool Not yet run jvpoulos/Attention-OCR/src/model/cnn.py
code served (permissive licence) · get_code("32711cdafa63eda4")
maybe_download Not yet run jvpoulos/Attention-OCR/src/model/data_utils.py
code served (permissive licence) · get_code("43ea785c6d00f3e4")
process_args Not yet run harvardnlp/im2markup/scripts/preprocessing/preprocess_images.py
code served (permissive licence) · get_code("ae08cb323bcbd595")
process_args Not yet run jvpoulos/Attention-OCR/src/launcher.py
code served (permissive licence) · get_code("e1ee189649199b8f")
rnn_decoder Not yet run jvpoulos/Attention-OCR/src/model/seq2seq.py
code served (permissive licence) · get_code("f44847e8b1c4bd91")
var_random Not yet run jvpoulos/Attention-OCR/src/model/cnn.py
code served (permissive licence) · get_code("213cd44559181629")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We present a neural encoder-decoder model to convert images into presentational markup based on a scalable coarse-to-fine attention mechanism. Our method is evaluated in the context of image-to-LaTeX generation, and we introduce a new dataset of real-world rendered mathematical expressions paired with LaTeX markup. We show that unlike neural OCR techniques using CTC-based models, attention-based approaches can tackle this non-standard OCR task. Our approach outperforms classical mathematical OCR systems by a large margin on in-domain rendered data, and, with pretraining, also performs well on out-of-domain handwritten data. To reduce the inference complexity associated with the attention-based approaches, we introduce a new coarse-to-fine attention layer that selects a support region before applying attention.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("1609.04938")
get_code_for_paper("1609.04938")
have("1609.04938")

Connect an agent — have() is free.