In this paper, we develop a novel method for view-based recognition of human action/activity from videos. By observing just a few frames, we can identify the activity that takes place in a video sequence. The basic idea of our method is that activities can be positively identified from a sparsely sampled sequence of a few body poses acquired from videos. In our approach, an activity is represented by a set of pose and velocity vectors for the major body parts (hands, legs, and torso) and stored in a set of multidimensional hash tables. We develop a theoretical foundation that shows that robust recognition of a sequence of body pose vectors can be achieved by a method of indexing and sequencing and it requires only a few pose vectors (i.e., sampled body poses in video frames). We find that the probability of false alarm drops exponentially with the increased number of sampled body poses. So, matching only a few body poses guarantees high probability for correct recognition. Our approach is parallel, i.e., all possible model activities are examined at one indexing operation since all of the model activities are stored in the same set of hash tables. In addition, our method is robust to partial occlusion since each body part is indexed separately. We use a sequence-based voting approach to recognize the activity invariant to the activity speed. Experiments performed with videos having eight different activities show robust recognition with our method. The method is also robust in conditions of varying view angle in the range of AE 30 degrees.

| As the first decade of the 21st century comes to a close, growth in multimedia delivery infrastructure and public demand for applications built on this backbone are converging like never before. The push towards reaching truly interactive multimedia technologies becomes stronger as our media consumption paradigms continue to change. In this paper, we will profile a technology leading the way in this revolution: active learning. Active learning is a strategy that helps alleviate challenges inherent in multimedia information retrieval through user interaction. We will show how active learning is ideally suited for the multimedia information retrieval problem by giving an overview of the paradigm and component technologies used with special attention given to the application scenarios in which these technologies are useful. Finally, we give insight into the future of this growing field and how it fits into the larger context of multimedia information retrieval.

Locality Sensitive Hash functions are invaluable tools for approximate near neighbor problems in high dimensional spaces. In this work, we are focused on LSH schemes where the similarity metric is the cosine measure. The contribution of this work is a new class of locality sensitive hash functions for the cosine similarity measure based on the theory of concomitants, which arises in order statistics. Con-Concomitant theory captures the relation between the order statistics of X and Y in the form of a rank distribution given by Prob(Rank(Yi)=j|Rank(Xi)=k). We exploit properties of the rank distribution towards developing a locality sensitive hash family that has excellent collision rate properties for the cosine measure.The computational cost of the basic algorithm is high for high hash lengths. We introduce several approximations based on the properties of concomitant order statistics and discrete transforms that perform almost as well, with significantly reduced computational cost. We demonstrate the practical applicability of our algorithms by using it for finding similar images in an image repository.

No abstract

Abstract.Recently, ranking and sorting problems have attracted the attention of researchers in the machine learning community. By ranking, we refer to categorizing examples into one of K categories. On the other hand, sorting refers to coming up with the ordering of the data that agrees with some ground truth preference function. As against standard approaches of treating ranking as a multiclass classification problem, in this paper we argue that ranking/sorting problems can be solved by exploiting the inherent structure present in data. We present efficient formulations that enable the use of standard binary classification algorithms to solve these problems, however the structure is still captured in our formulations. We further show that our approach subsumes the various approaches that were developed in the past. We evaluate our algorithm on both synthetic datasets and for a real world image processing problem. The results obtained demonstrate the superiority of our algorithm over multiclass classification and other similar approaches for ranking/sorting data.