Showing papers by "Gaurav Harit published in 2015"

Document Retrieval with Unlimited Vocabulary

Abstract: In this paper, we describe a classifier based retrieval scheme for efficiently and accurately retrieving relevant documents. We use SVM classifiers for word retrieval, and argue that the classifier based solutions can be superior to the OCR based solutions in many practical situations. We overcome the practical limitations of the classifier based solution in terms of limited vocabulary support, and availability of training data. In order to overcome these limitations, we design a one-shot learning scheme for dynamically synthesizing classifiers. Given a set of SVM classifiers, we appropriately join them to create novel classifiers. This extends the classifier based retrieval paradigm to an unlimited number of classes (words) present in a language. We validate our method on multiple datasets, and compare it with popular alternatives like OCR and word spotting. Even on a language like English, where OCRs have been fairly advanced, our method yields comparable or even superior results. Our results are significant since we do not use any language specific post-processing for obtaining this performance. For better accuracy of the retrieved list, we use query expansion. This also allows us to seamlessly adapt our solution to new fonts, styles and collections.

Domain adaptation by aligning locality preserving subspaces

Abstract: The mismatch between the training data and the test data distributions is a challenging issue while designing many practical computer vision systems. In this paper, we propose a domain adaptation technique to tackle this issue. We are interested in a domain adaptation scenario where source domain has large amount of labeled examples and the target domain has large amount of unlabeled examples. We align the source domain subspace with the target domain subspace in order to reduce the mismatch between the two distributions. We model the subspace using Locality Preserving Projections (LPP). Unlike previous subspace alignment approaches, we introduce a strategy to effectively utilize the training labels in order to learn discriminative subspaces. We validate our domain adaptation approach by testing it on two different domains, i.e. handwritten and printed digit images. We compare our approach with other existing approaches and show the superiority of our method.

Gradient sensitive kernel for Image Denoising, using Gaussian Process Regression

Abstract: We target the problem of Image Denoising using Gaussian Processes Regression (GPR). Being a non-parametric regression technique, GPR has received much attention in the recent past and here we further explore its versatility by applying it to a denoising problem. The focus is primarily on the design of a local gradient sensitive kernel that captures pixel similarity in the context of image denoising. This novel kernel formulation is used to shape the smoothness of the joint GP prior. We apply the GPR denoising technique to small patches and then stitch back these patches, this allows the priors to be local and relevant, also this helps us in dealing with GPR complexity. We demonstrate that our GPR based technique gives better PSNR values in comparison to existing popular denoising techniques.

Segmenting printed text and handwritten annotation by Spectral Partitioning

Abstract: This paper addresses the problem of segmenting handwritten annotations on scientific research papers. The motivation of this work is to geometrically segment the complex cases of handwritten annotations, including marks, cuts and special symbols along, with the regular text. Our work particularly focuses on documents that have multi-oriented handwritten [1] annotations rather than annotations in controlled scenario [2]. Spectral Partitioning is adopted as the segmentation scheme to separate the printed text and annotations. A new feature Envelope Straightness is developed and included in our feature set. This leads to an improvement of accuracy over the state-of-the-art features. The experiments are performed on two datasets: 40 documents authored by two writers from IAM dataset, comprising only printed and handwritten text and a self created dataset of 40 scientific papers from various proceedings annotated by a reader, comprising varied types of annotations. In the framework of spectral partitioning, our feature set has achieved a recall of 98.39% for printed text and precision of 85.40% for handwritten annotations on our dataset. For IAM dataset our feature set has achieved a recall of 81.89% for printed text and a precision of 69.67% for handwritten annotations. The results achieved on both dataset are better compared with results obtained using [3] [1].