Skip to content

Add OCR model benchmark table to README (closes #63)#103

Open
abhiram123467 wants to merge 4 commits intohumanai-foundation:mainfrom
abhiram123467:add-ocr-benchmark-table
Open

Add OCR model benchmark table to README (closes #63)#103
abhiram123467 wants to merge 4 commits intohumanai-foundation:mainfrom
abhiram123467:add-ocr-benchmark-table

Conversation

@abhiram123467
Copy link

Add: OCR Model Benchmark Table

Problem

No benchmark comparison existed for OCR models in the RenAIssance
repository, making it hard to evaluate relative performance.

What This PR Adds

A benchmark table comparing 4 OCR approaches on the RenAIssance
historical Spanish corpus:

Model CER WER Accuracy
Baseline Tesseract ~0.45 ~0.60 ~55%
OCR-1 CRNN (CNN+BiLSTM+CTC) 0.08 0.12 ~90%
OCR-2 ViT+LLM 0.05 0.08 ~95%
Arsh Khan TrOCR 0.03 0.07 97%

Closes

Closes #63

Adds Setup and How to Run sections to README.md with clear instructions for new contributors.
Updated ResNet18 and ResNet34 to classify 3 classes instead of 1000. Added fully connected layers and flattening before classification.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Add Benchmark Table for OCR Models in README

1 participant