abstract
- Dataset Summary The PatFigVQA dataset is introduced in the paper Patent Figure Classification using Large Vision-language Models accepted at ECIR 2025. The dataset is designed specifically for fine-tuning and evaluating Large Vision-language Models (LVLMs) in few-shot learning setting across multiple aspects, including type, projection, objects and USPC class. The PatFigVQA dataset is used alongside another dataset called PatFigCLS, which is intended for patent figure classification and evaluation. The dataset is sourced from two exisiting datasets: Extended CLEF-IP 2011, and DeepPatent2 Data Format The dataset is stored in .tar files for fast and efficient read access. Data Fields __key__: unique sample id image.png: patent figure file task.txt: type of question (e.g. binary, multiple-choice or open-ended) question.txt: natural language question asking about the concept depicted in figure answer.txt: textual answer for the question concept.txt: concept depicted in the patent figure for the given aspect Data Splits For each aspect, multiple data splits exist: train_9, train_18, train_27, train_54, train_81, train_150, val and test. How to Use The recommended approach is using the Python library `webdataset`. Below is an example code. import io from PIL import Image from torchvision.transforms import Compose, ToTensor import webdataset as wds from braceexpand import braceexpand def transform(image): return Compose([ToTensor()])(image) dataset = ( wds.WebDataset( braceexpand('PatFigVQA/object/train_150/shard-{000000..000042}.tar'), shardshuffle=1000, ) .shuffle(1000) .to_tuple( '__key__', 'image.png', 'concept.txt', 'task.txt', 'question.txt', 'answer.txt', ) .map_tuple( lambda key: key, lambda image: transform(Image.open(io.BytesIO(image))), lambda concept: concept.decode('utf-8'), lambda task: task.decode('utf-8'), lambda question: question.decode('utf-8'), lambda answer: answer.decode('utf-8'), ) ) dataloder = wds.WebLoader(dataset) Source Code The source code used to produce this dataset can be found at https://github.com/TIBHannover/patent-figure-classification Licensing Information PatFigCLS dataset is released under GNU General Public License v3.0.