Optical Chemical Structure Recognition

GTR-CoT uses graph traversal chain-of-thought reasoning to improve optical chemical structure recognition accuracy.

MolRec achieves 95%+ accuracy on simple structures but struggles with complex diagrams, revealing rule-based OCSR limits …

MolRec achieves 95% accuracy on 1000 molecular diagrams at TREC 2011 with detailed OCR and structure reconstruction.

Novel OCSR method creating molecular fingerprints from images through functional group segmentation for database …

αExtractor uses ResNet-Transformer to extract chemical structures from literature images, including noisy and hand-drawn …

Fujiyoshi et al.'s segment-based approach for recognizing chemical structures in challenging Japanese patent images with …

Two-stage CNN approach for converting molecular images to SMILES using CDDD embeddings and extensive data augmentation.

Chen et al.'s dual-stream encoder approach for robust molecular structure recognition from diverse real-world images …

Filippov & Nicklaus's open-source rule-based system for converting molecular structure images into machine-readable …

MolParser-7M is the largest OCSR dataset with 7.7M image-text pairs of molecules and E-SMILES, including 400k real-world …

MolParser converts molecular images from scientific documents to machine-readable formats using end-to-end learning.

Campos & Ji's method for converting 2D molecular images to SMILES strings using Transformers and SELFIES representation.