In this paper, we propose a novel multiple pyramids based image inpainting method using local patch statistics and geometric feature-based sparse representation to maintain texture consistency and structure coherence. First, we approximate each patch in the target region (region to be inpainted) by statistically dominant local candidate patches to preserve local consistency. Then each approximated patch is refined by a sparse representation of candidate patches based on local steering kernel (LSK) feature to retain texture quality. We also propose a multiple pyramids based approach to generate several inpainted versions of the input image, one for each of the pyramids. Finally, we combine the inpainted images by gradient-based weighted average to produce the final inpainted image. This approach helps to maintain structure coherence and to remove artifacts which may appear in the inpainted images due to different initial scales of the individual pyramids. The proposed method is tested on a wide range of natural images for scratch and blob/object removal. We have presented both quantitative and qualitative comparison with the existing methods to demonstrate the superiority of the proposed method.

Humans can understand event roles like agent and patient within videos of simple shapes moving around just using simple motion heuristics. As existing computational systems do not directly address human understanding of these events, we develop the first computational model that can simulate human performance in these tasks. We develop an approach heuristic that can simulate how human recognition of chasing is influenced by the angle that the chaser uses to approach the chasee. We also created a causality heuristic that captures human sensitivity to contact between the pusher and the pushee, as well as a delay in launching. Careful modelling of psychological studies of infants and adults behaviour can yield insights that may enhance computational systems for action understanding.

This study presents a novel ensemble classifier‐based off‐line handwritten word recognition system following a holistic approach. Here each handwritten word is recognised using two handcrafted features, namely (i) Arnold transform‐based feature that addresses local directional feature which depends on the stroke orientation distribution of cursive word and (ii) oriented curvature‐based feature which is basically the histogram of curvelet index and one machine generated feature using deep convolution neural network (DCNN). In this study, a new architecture of DCNN is proposed for handwritten word recognition. These features are recognised by three classifiers separately. Finally, the decision of three classifiers is combined to predict the ultimate word class level. To fuse the decision of individual classifiers, the authors have explored three strategies: (i) vote for strongest decision, (ii) vote for majority decision and (iii) vote for the sum of the decisions. The proposed handwritten word recognition system is tested on three handwritten word databases: (i) CENPARMI database, (ii) IAM database and (iii) ISIHWD database. The performance of the proposed system is promising and comparable to state‐of‐the‐art handwriting recognition systems.

Due to occlusion, lighting condition, variation in clothing dance video classification is a challenging problem in computer vision domain. In this paper we present a local spatiotemporal feature model on manifold for Indian Classical Dance (ICD) classification. We represent features at each space-time interest point as a covariance matrix by fusing different order spatial and temporal derivatives. Each video clip is then represented in bag-of-words framework on manifold using Jensen-Bregman LogDet Divergence. Classification is done by popular non-linear SVM with ?2-kernel. We evaluate our system on a ICD dataset created from YouTube and get 69.39% accuracy which is better than that of the state-of-the-art human activity classification algorithms. We have also tested our algorithms on human activity benchmark datasets like KTH, and UCF50 and get promising results compared to the state-of-the-art methods.

This paper presents a novel space-time feature-based human activity analysis system. We detect Space Time Interest Points (STIP) and generate their description based on the facet model. The proposed approach detects interest points in video data using the three-dimensional facet model efficiently. Then we describe each interest point by three-dimensional Haar wavelet transform and time derivatives of different order obtained from said facet model. Here we represent each video clip following the bag-of-words approach by learning feature specific dictionary. Finally, classification is done using non-linear SVM with χ2-kernel. We evaluate the performance of our system on standard datasets like Weizmann, KTH, UCF sports, ICD, UCF YouTube, and UCF50 and get better, or at least comparable results compared to other state-of-the-art systems.

In this paper, we address an interesting application of computer vision technique, namely classification of Indian Classical Dance (ICD). With the best of our knowledge, the problem has not been addressed so far in computer vision domain. To deal with this problem, we use a sparse representation based dictionary learning technique. First, we represent each frame of a dance video by a pose descriptor based on histogram of oriented optical flow (HOOF), in a hierarchical manner. The pose basis is learned using an on-line dictionary learning technique. Finally each video is represented sparsely as a dance descriptor by pooling pose descriptor of all the frames. In this work, dance videos are classified using support vector machine (SVM) with intersection kernel. Our contribution here are two folds. First, to address dance classification as a new problem in computer vision and second, to present a new action descriptor to represent a dance video which overcomes the problem of the “Bags-of-Words” model. We have tested our algorithm on our own ICD dataset created from the videos collected from YouTube. An accuracy of 86.67% is achieved on this dataset. Since we have proposed a new action descriptor too, we have tested our algorithm on well known KTH dataset. The performance of the system is comparable to the state-of-the-art.