Tiny CNN for feature point description for document analysis: approach and dataset
This work addresses the need for efficient feature matching in document analysis on resource-limited devices, though it is incremental as it adapts existing lightweight CNN methods to a specific domain.
The authors tackled feature point description for document analysis by constructing a specialized dataset and training a lightweight CNN, achieving improved performance on document datasets MIDV-500 and MIDV-2019 compared to using general datasets like HPatches.
In this paper, we study the problem of feature points description in the context of document analysis and template matching. Our study shows that the specific training data is required for the task especially if we are to train a lightweight neural network that will be usable on devices with limited computational resources. In this paper, we construct and provide a dataset with a method of training patches retrieval. We prove the effectiveness of this data by training a lightweight neural network and show how it performs in both documents and general patches matching. The training was done on the provided dataset in comparison with HPatches training dataset and for the testing we use HPatches testing framework and two publicly available datasets with various documents pictured on complex backgrounds: MIDV-500 and MIDV-2019.