MolRec: Chemical Structure Recognition at CLEF 2012
MolRec achieves 95%+ accuracy on simple structures but struggles with complex diagrams, revealing rule-based OCSR limits …
Optical Chemical Structure Recognition
Notes on methods for extracting molecular structures from images. Covers OCR-based, deep learning, and vision-language model approaches.
MolRec: Rule-Based OCSR System
Rule-based system for optical chemical structure recognition using vectorization and geometric analysis, achieving 95% …
SubGrapher: Visual Fingerprinting of Chemical Structures
Novel OCSR method creating molecular fingerprints from images through functional group segmentation for database …
αExtractor: Chemical Info from Biomedical Literature
αExtractor uses ResNet-Transformer to extract chemical structures from literature images, including noisy and hand-drawn …
ChemInfty: Chemical Structure Recognition in Patent Images
Fujiyoshi et al.'s segment-based approach for recognizing chemical structures in challenging Japanese patent images with …
MolNexTR: Dual-Stream Molecular Image Recognition
Dual-stream encoder combining ConvNext and ViT for robust optical chemical structure recognition across diverse …
MolParser-7M & WildMol: Large-Scale OCSR Datasets
MolParser-7M is the largest OCSR dataset with 7.7M image-text pairs of molecules and E-SMILES, including 400k real-world …
MolParser: End-to-End Molecular Structure Recognition
MolParser converts molecular images from scientific documents to machine-readable formats using end-to-end learning with …