Abstract

Automatic annotation of an audio or a music piece with multiple labels helps in understanding the composition of a music. Such meta-level information can be very useful in applications such as music transcription, retrieval, organization and personalization. In this work, we formulate the problem of annotation as multi-label classification which is considerably different from that of a popular single (binary or multi-class) label classification. We employ both the nearest neighbour and max-margin (SVM) formulations for the automatic annotation. We consider K-NN and SVM that are adapted for multi-label classification using one-vs-rest strategy and a direct multi-label classification formulation using ML-KNN and M3L. In the case of music, often the signatures of the labels (e.g. instruments and vocal signatures) are fused in the features. We therefore propose a simple feature augmentation technique based on non …

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 …

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 …

Abstract

The dynamic time warping (DTW) distance is a popular similarity measure for comparing time series data. It has been successfully applied in many fields like speech recognition, data mining and information retrieval to automatically cope with time deformations and variations in the length of the time dependent data. There have been attempts in the past to define kernels on DTW distance. These kernels try to approximate the DTW distance. However, these have quadratic complexity and these are computationally expensive for large time series. In this paper, we introduce FastDTW kernel, which is a linear approximation of the DTW kernel and can be used with linear SVM.

Abstract

Human pose as a query modality is an alternative and rich experience for image and video retrieval. We present a novel approach for the task of human pose retrieval, and make the following contributions: first, we introduce `deep poselets' for pose-sensitive detection of various body parts, that are built on convolutional neural network (CNN) features. These deep poselets significantly outperform previous instantiations of Berkeley poselets [2]. Second, using these detector responses, we construct a pose representation that is suitable for pose search, and show that pose retrieval performance exceeds previous methods by a factor of two. The compared methods include Bag of visual words [24], Berkeley poselets [2] and Human pose estimation algorithms [28]. All the methods are quantitatively evaluated on a large dataset of images built from a number of standard benchmarks together with frames from Hollywood …

Abstract

Traditional visual servoing is able to navigate a robotic system between two views of the same object. However, it is not designed to servo between views of different objects. In this paper, we consider a novel problem of servoing any instance (exemplar) of an object category to a desired pose (view) and propose a strategy to accomplish the task. We use features that semantically encode the locations of object parts and define the servoing error as the difference between positions of corresponding parts in the image space. Our controller is based on the linear combination of 3D models, such that the resulting model interpolates between the given and desired instances. We conducted our experiments on five different object categories in simulation framework and show that our approach achieves the desired pose with smooth trajectory. Furthermore, we show the performance gain achieved by using a linear …

Abstract

In this work, we propose an online handwriting solution, where the data is captured with the help of depth sensors. Users may write in the air and our method recognizes it in real time using the proposed feature representation. Our method uses an efficient fingertip tracking approach and reduces the necessity of pen-up/pen-down switching. We validate our method on two depth sensors, Kinect and Leap Motion Controller. On a dataset collected from 20 users, we achieve a recognition accuracy of 97.59% for character recognition. We also demonstrate how this system can be extended for lexicon recognition with reliable performance. We have also prepared a dataset containing 1,560 characters and 400 words with the intention of providing common benchmark for handwritten character recognition using depth sensors and related research.

Abstract

Most of the popular optical character recognition (OCR) architectures use a set of handcrafted features and a powerful classifier for isolated character classification. Success of these methods often depend on the suitability of these features for the language of interest. In recent years, whole word recognition based on Recurrent Neural Networks (RNN) has gained popularity. These methods use simple features such as raw pixel values or profiles. Success of these methods depend on the learning capabilities of these networks to encode the script and language information. In this work, we investigate the possibility of learning an appropriate set of features for designing OCR for a specific language. We learn the language specific features from the data with no supervision. This enables the seamless adaptation of the architecture across languages. In this work, we learn features using a stacked Restricted Boltzman …

Abstract

Dynamic time warping (DTW) is a popular distance measure used for recognition free document image retrieval. However, it has quadratic complexity and hence is computationally expensive for large scale word image retrieval. In this paper, we use a fast approximation to the DTW distance, which makes word retrieval efficient. For a pair of sequences, to compute their DTW distance, we need to find the optimal alignment from all the possible alignments. This is a computationally expensive operation. In this work, we learn a small set of global principal alignments from the training data and avoid the computation of alignments for query images. Thus, our proposed approximation is significantly faster compared to DTW distance, and gives 40 times speed up. We approximate the DTW distance as a sum of multiple weighted Eulidean distances which are known to be amenable to indexing and efficient retrieval. We …

Abstract

In this work, we investigate the utility of Recurrent Neural Networks (RNNs) for script and language identification. Both these problems have been attempted in the past with representations computed from the distribution of connected components or characters (e.g. texture, n-gram). Often these features are computed from a larger segment (a paragraph or a page). We argue that one can predict the script or language with minimal evidence (e.g. given only a word or a line) very accurately with the help of a pre-trained RNN. We propose a simple and generic solution for the task of script and language identification which do not require any special tuning. Our method represents the word images as a sequence of feature vectors, and employ the RNNs for the identification. We verify the method on a large corpus of more than 15.03M words from 55K document images comprising 15 scripts and languages. We report an …