A significant fraction of information searches are motivated by the user’s primary task . An ideal search engine would be able to use information captured from the primary task to proactively retrieve useful information. Previous work has shown that many information retrieval activities depend on the primary task in which the retrieved information is to be used, but fairly little research has been focusing on methods that automatically learn the informational intents from the primary task context. We study how the implicit primary task context can be used to model the user’s search intent and to proactively retrieve relevant and useful information. Data comprising of logs from a user study, in which users are writing an essay, demonstrate that users’ search intents can be captured from the task and relevant and useful information can be proactively retrieved. Data from simulations with several datasets of different complexity show that the proposed approach of using primary task context generalizes to a variety of data. Our findings have implications for the design of proactive search systems that can infer users’ search intent implicitly by monitoring users’ primary task activities.

Gesture recognition using RGB-D sensors has currently an important role in many fields such as human-computer interfaces, robotics control, and sign language recognition. However, the recognition of hand gestures under natural conditions with low spatial resolution and strong motion blur still remains an open research question. In this paper we propose an online gesture recognition method for multimodal RGB-D data. We extract multiple hand features with the assistance of body and hand masks from RGB and depth frames, and full-body features from the skeleton data. These features are classified by multiple Extreme Learning Machines on the frame level. The classifier outputs are then modeled on the sequence level and fused together to provide the final classification results for the gestures. We apply our method on the ChaLearn 2013 gesture dataset consisting of natural signs with the hand diameters in the images around 20-40 pixels. Our method achieves an 85% recognition accuracy with 20 gesture classes and can perform the recognition in real-time.

Convolutional neural networks have recently been used to obtain record-breaking results in many vision benchmarks. In addition, the intermediate layer activations of a trained network when exposed to new data sources have been shown to perform very well as generic image features, even when there are substantial differences between the original training data of the network and the new domain. In this paper, we focus on scene recognition and show that convolutional networks trained on mostly object recognition data can successfully be used for feature extraction in this task as well. We train a total of four networks with different training data and architectures, and show that the proposed method combining multiple scales and multiple features obtains state-of-the-art performance on four standard scene datasets.

As the growth of stored personal digital information, such as photographs and emails, is continuously increasing, new tools for browsing and searching are needed. We introduce an intelligent mobile information access tool for personal data. The data are presented in an adaptive timeline where the displayed items function as search cues. The novelty is that the visualization is dynamically changed to emphasize relevant items, which makes them easier to recognize and select. The relevance is inferred during usage of the system from user feedback. In a user study, the dynamic timeline-based interface on a mobile device was shown to require less effort than conventional textual search.

Gesture recognition is needed in many applications such as human-computer interaction and sign language recognition. The challenges of building an actual recognition system do not lie only in reaching an acceptable recognition accuracy but also with requirements for fast online processing. In this paper, we propose a method for online gesture recognition using RGB-D data from a Kinect sensor. Frame-level features are extracted from RGB frames and the skeletal model obtained from the depth data, and then classified by multiple extreme learning machines. The outputs from the classifiers are aggregated to provide the final classification results for the gestures. We test our method on the ChaLearn multi-modal gesture challenge data. The results of the experiments demonstrate that the method can perform effective multi-class gesture recognition in real-time.

Many emerging application areas in video and image processing require large-scale visual concept detection. Examples include content-based indexing of online user-generated videos and 24/7 archival of TV broadcasts. The current state of the art in concept detection uses bag-of-visual-words features with computationally heavy exponential kernel classifiers. We argue that this classifier approach is not feasible for large-scale real-time applications, and propose instead to use combinations of approximate additive kernel classifiers. By using explicit kernel maps and the power mean SVM, followed by fusion of classifiers trained on different features, we achieve high retrieval precision while retaining real-time performance for large sets of concepts. This paper presents a series of experiments with the large-scale TRECVID 2012 video database and the commonly used Fifteen Scene Categories image database. We show significantly improved retrieval performance over standard linear classifiers, and by late fusion over several visual features, the approximative additive kernels outperform any single exponential kernel in only a fraction of the detection time.

The recent development of large-scale multimedia concept ontologies has provided a new momentum for research in the semantic analysis of multimedia repositories. Different methods for generic concept detection have been extensively studied, but the question of how to exploit the structure of a multimedia ontology and existing inter-concept relations has not received similar attention. In this paper, we present a clustering-based method for modeling semantic concepts on low-level feature spaces and study the evaluation of the quality of such models with entropy-based methods. We cover a variety of methods for assessing the similarity of different concepts in a multimedia ontology. We study three ontologies and apply the proposed techniques in experiments involving the visual and semantic similarities, manual annotation of video, and concept detection. The results show that modeling inter-concept relations can provide a promising resource for many different application areas in semantic multimedia processing.

In this paper we present a clustering-based method for representing semantic concepts on multimodal low-level feature spaces and study the evaluation of the goodness of such models with entropy-based methods. As different semantic concepts in video are most accurately represented with different features and modalities, we utilize the relative model-wise confidence values of the feature extraction techniques in weighting them automatically. The method also provides a natural way of measuring the similarity of different concepts in a multimedia lexicon. The experiments of the paper are conducted using the development set of the TRECVID 2005 corpus together with a common annotation for 39 semantic concepts

Development of content-based image retrieval (CBIR) techniques has suffered from the lack of standardized ways for describing visual image content. Luckily, the MPEG-7 international standard is now emerging as both a general framework for content description and a collection of specific agreed-upon content descriptors. We have developed a neural, self-organizing technique for CBIR. Our system is named PicSOM and it is based on pictorial examples and relevance feedback (RF). The name stems from "picture" and the self-organizing map (SOM). The PicSOM system is implemented by using tree structured SOMs. In this paper, we apply the visual content descriptors provided by MPEG-7 in the PicSOM system and compare our own image indexing technique with a reference system based on vector quantization (VQ). The results of our experiments show that the MPEG-7-defined content descriptors can be used as such in the PicSOM system even though Euclidean distance calculation, inherently used in the PicSOM system, is not optimal for all of them. Also, the results indicate that the PicSOM technique is a bit slower than the reference system in starting to find relevant images. However, when the strong RF mechanism of PicSOM begins to function, its retrieval precision exceeds that of the reference system.

Content-based image retrieval is an important approach to the problem of processing the increasing amount of visual data. It is based on automatically extracted features from the content of the images, such as color, texture, shape and structure. We have started a project to study methods for content-based image retrieval using the self-organizing map (SOM) as the image similarity scoring method. Our image retrieval system, named PicSOM, can be seen as a SOM-based approach to relevance feedback which is a form of supervised learning to adjust the subsequent queries based on the user's responses during the information retrieval session. In PicSOM, a separate tree structured SOM (TS-SOM) is trained for each feature vector type in use. The system then adapts to the user's preferences by returning her more images from those SOMs where her responses have been most densely mapped.