Use Case and High-Level Description

This is an im2latex composite model that recognizes latex formulas. The model uses vocabulary file to predict sequence of latex tokens. The model is built on the resnext50 backbone with additional attention-based text recognition head. The model was trained on internal Intel's dataset containing images of handwritten polynomial equations. The equations consist of tokens from the corresponding to this model vocabulary file.

Vocabulary file is distributed with the model. Model can predict letters, numbers and upperscript.

Example of the input data

Example of the output

- 4 . 6 c ^ { 2 } d ^ { - 6 0 }

Composite model specification

Im2latex-match-images metric is calculated by this algorithm

Encoder model specification

The formula-recognition-polynomials-handwritten-0001-encoder model is a ResNeXt-50 like backbone with some initialization layers for decoder

Inputs

1. Name: imgs , shape: [1x3x96x990]. An input image in the [1xCxHxW] format. The expected channel order is BGR.

Outputs

1. Name: hidden, shape: [1x512]. Initial context state of the LSTM cell.
2. Name: context, shape: [1x512]. Initial hidden state of the LSTM cell.
3. Name: init_0, shape: [1x256]. Initial state of the decoder.
4. Name: row_enc_out, shape: [1x6x62x512]. Features from encoder that are fed to a decoder.

Decoder model specification

The formula-recognition-polynomials-handwritten-0001-decoder model is an LSTM based decoder with attention module.

Inputs

1. Name: dec_st_c , shape: [1x512]. Current context state of the LSTM cell.
2. Name: dec_st_h , shape: [1x512]. Current hidden state of the LSTM cell.
3. Name: output_prev, shape: [1x256]. Current state of the decoder.
4. Name: row_enc_out, shape: [1x6x62x512]. Encoded features.
5. Name: tgt, shape: [1x1]. Index of the previous symbol.

Outputs

1. Name: dec_st_c , shape: [1x512]. Current context state of the LSTM cell.
2. Name: dec_st_h , shape: [1x512]. Current hidden state of the LSTM cell.
3. Name: output, shape: [1x256]. Current state of the decoder.
4. Name: logit, shape: [1xN], where N is a vocabulary size. Classification confidence scores in the [0, 1] range for every token.

Legal Information

[*] Other names and brands may be claimed as the property of others.