Abstract

We focus on the problem of wearer's action recognition in first person aka egocentric videos. This problem is more challenging than third person activity recognition due to unavailability of wearer's pose and sharp movements in the videos caused by the natural head motion of the wearer. Carefully crafted features based on hands and objects cues for the problem have been shown to be successful for limited targeted datasets. We propose convolutional neural networks (CNNs) for end to end learning and classification of wearer's actions. The proposed network makes use of egocentric cues by capturing hand pose, head motion and saliency map. It is compact. It can also be trained from relatively small number of labeled egocentric videos that are available. We show that the proposed network can generalize and give state of the art performance on various disparate egocentric action datasets.

Abstract

Crowded urban environments are composed of different types of dynamic and static elements. Learning and classification of features is a major task in solving the localization problem in such environments. This work presents a gradual learning methodology to learn the useful features using multiple experiences. The usefulness of an observed element is evaluated by a scoring mechanism which uses two scores – reliability and distinctiveness. The visual features thus learned are used to partition the visual map into smaller regions. The robot is efficiently localized in such a partitioned environment using two-level localization. The concept of active map (AM) is proposed here, which is a map that represents one partition of the environment in which there is a high probability of the robot existing. High-level localization is used to track the mode of the AMs using discrete Bayes filter. Low-level localization uses a bag-of …

Abstract

We propose a semi-supervised self-training method for fine-tuning human pose estimations in videos that provides accurate estimations even for complex sequences. We surpass state-of-the-art on most of the datasets used and also show a 2.33% gain over the baseline on our new dataset of unrestricted sports videos. The self-training model presented has two components: a static Pictorial Structure (PS) based model and a dynamic ensemble of exemplars. We present a pose quality criteria that is primarily used for batch selection and automatic parameter selection. The same criteria works as a low-level pose evaluator used in post-processing. We set a new challenge by introducing a full human body-parts annotated complex dataset, CVIT-SPORTS, which contains complex videos from the sports domain. The strength of our method is demonstrated by adapting to videos of complex activities such as cricket …

Abstract

Several methods have been proposed for large-scale 3D reconstruction from large, unorganized image collections. A large reconstruction problem is typically divided into multiple components which are reconstructed independently using structure from motion (SFM) and later merged together. Incremental SFM methods are most popular for the basic structure recovery of a single component. They are robust and effective but strictly sequential in nature. We present a multistage approach for SFM reconstruction of a single component that breaks the sequential nature of the incremental SFM methods. Our approach begins with quickly building a coarse 3D model using only a fraction of features from given images. The coarse model is then enriched by localizing remaining images and matching and triangulating remaining features in subsequent stages. The geometric information available in form of the coarse model allows us to make these stages effective, efficient, and highly parallel. We show that our method produces similar quality models as compared to standard SFM methods while being notably fast and parallel.

Abstract

Accurately annotated large video data is critical for the development of reliable surveillance and automotive related vision solutions. In this work, we propose an efficient and yet accurate annotation scheme for objects in videos (pedestrians in this case) with minimal supervision. We annotate objects with tight bounding boxes. We propagate the annotations across the frames with a self training based approach. An energy minimization scheme for the segmentation is the central component of our method. Unlike the popular grab cut like segmentation schemes, we demand minimal user intervention. Since our annotation is built on an accurate segmentation, our bounding boxes are tight. We validate the performance of our approach on multiple publicly available datasets.

Abstract

Short internet video clips like vines present a significantly wild distribution compared to traditional video datasets. In this paper, we focus on the problem of unsupervised action classification in wild vines using traditional labeled datasets. To this end, we use a data augmentation based simple domain adaptation strategy. We utilize semantic word2vec space as a common subspace to embed video features from both, labeled source domain and unlabled target domain. Our method incrementally augments the labeled source with target samples and iteratively modifies the embedding function to bring the source and target distributions together. Additionally, we utilize a multi-modal representation that incorporates noisy semantic information available in form of hash-tags. We show the effectiveness of this simple adaptation technique on a test set of vines and achieve notable improvements in performance.

Abstract

Recognizing scene text is a challenging problem, even more so than the recognition of scanned documents. This problem has gained significant attention from the computer vision community in recent years, and several methods based on energy minimization frameworks and deep learning approaches have been proposed. In this work, we focus on the energy minimization framework and propose a model that exploits both bottom-up and top-down cues for recognizing cropped words extracted from street images. The bottom-up cues are derived from individual character detections from an image. We build a conditional random field model on these detections to jointly model the strength of the detections and the interactions between them. These interactions are top-down cues obtained from a lexicon-based prior, i.e., language statistics. The optimal word represented by the text image is obtained by minimizing the …

Abstract

We present an approach for automatically identifying the script of the text localized in the scene images. Our approach is inspired by the advancements in mid-level features. We represent the text images using mid-level features which are pooled from densely computed local features. Once text images are represented using the proposed mid-level feature representation, we use an off-the-shelf classifier to identify the script of the text image. Our approach is efficient and requires very less labeled data. We evaluate the performance of our method on a recently introduced CVSI dataset, demonstrating that the proposed approach can correctly identify script of 96.70% of the text images. In addition, we also introduce and benchmark a more challenging Indian Language Scene Text (ILST) dataset for evaluating the performance of our method.

Abstract

User-given tags or labels are valuable resources for semantic understanding of visual media such as images and videos. Recently, a new type of labeling mechanism known as hashtags have become increasingly popular on social media sites. In this paper, we study the problem of generating relevant and useful hash-tags for short video clips. Traditional data driven approaches for tag enrichment and recommendation use direct visual similarity for label transfer and propagation. We attempt to learn a direct low-cost mapping from video to hash-tags using a two step training process. We first employ a natural language processing (NLP) technique, skip-gram models with neural network training to learn a low-dimensional vector representation of hash-tags (Tag2Vec) using a corpus of ∼ 10 million hash-tags. We then train an embedding function to map video features to the low-dimensional Tag2vec space. We learn this embedding for 29 categories of short video clips with hash-tags. A query video without any tag-information can then be directly mapped to the vector space of tags using the learned embedding and relevant tags can be found by performing a simple nearest-neighbor retrieval in the Tag2Vec space. We validate the relevance of the tags suggested by our system qualitatively and quantitatively with a user study.

Abstract

In Indian scenario, a document analysis system has to support multiple languages at the same time. With emerging multilingualism in urban India, often bilingual, trilingual or even more languages need to be supported. This demands development of a multilingual OCR system which can work seamlessly across Indic scripts. In our approach the script is identified at word level, prior to the recognition of the word. An end-to-end RNN based architecture which can detect the script and recognize the text in a segmentation-free manner is proposed for this purpose. We demonstrate the approach for 12 Indian languages and English. It is observed that, even with the similar architecture, performance on Indian languages are poorer compared to English. We investigate this further. Our approach is evaluated on a large corpus comprising of thousands of pages. The Hindi OCR is compared with other popular OCRs for the …