This paper proposes a method for retrieving the original information X out of multiple sets of side information (SI) such as Y1, Y2, Y3, etc. These sets are generated using key frames that are encoded by a conventional video codec. By using the method of majority rule, the most reliable bits from the decoded Yi’s are selected for the composition of Y∞, the conditional entropy H(X|Y∞) of which b...

In this paper, we propose an RGB-D visual odometry method that both minimizes the photometric error and aligns the edges between frames. The combination of the direct photometric information [1] and the edge features leads to higher tracking accuracy and enables the approach to deal with challenging texture-less scenes. In contrast to traditional line feature based methods [2], we involve all e...

We propose a system capable in real time of adding controllable and plausible oscillating physical like reaction effects in response to external forces (perturbations). These oscillating effects may be used to modify a motion or to customize it in a cartoon like way. The core of our system is based on several connected 3D pendulums with a propagating reaction. These pendulums always return to a...

We present a novel method to measure saliency in molecular dynamics simulation data. This saliency measure is based on a multiscale center-surround mechanism, which is fast and efficient to compute. We explore the use of the saliency function to guide the selection of representative and anomalous timesteps for summarization of simulations. To this end, we also introduce a multiscale keyframe se...

In this paper we focus on a new problem of event coreference resolution across television news videos. Based on the observation that the contents from multiple data modalities are complementary, we develop a novel approach to jointly encode effective features from both closed captions and video key frames. Experiment results demonstrate that visual features provided 7.2% absolute F-score gain o...

This paper presents a framework for generating smooth-looking transformations between pairs of surfaces that may differ in topology. The user controls the transformation by specifying a sparse control mesh on each surface and by associating each face in one control mesh with a corresponding face in the other. The algorithm builds a transformation from this information in two steps. The first st...

Non-Identical Duplicate video detection is a challenging research problem. Non-Identical Duplicate video are a pair of videos that are not exactly identical but are almost similar. In this paper, we evaluate two methods Keyframe-based and Tomography-based methods to determine the Non-Identical Duplicate videos. These two methods make use of the existing scale based shift invariant (SIFT) method...

This paper presents a new approach for human walking modeling from monocular image sequences. A kinematics model and a walking motion model are introduced in order to exploit prior knowledge. The proposed technique consists of two steps. Initially, an efficient feature point selection and tracking approach is used to compute feature points’ trajectories. Peaks and valleys of these trajectories ...

This paper describes the participation of MIRACLE research consortium at the VideoCLEF track at CLEF 2008. We took part in both the main mandatory Classification task that consists in classifying videos of television episodes using speech transcripts and metadata, and the Keyframe Extraction task, whose objective is to select keyframes that represent individual episodes from a set of supplied k...

We propose a new method for extracting key frames from a motion capture sequence. Our proposed approach consists of two steps. In the first step, we propose a new metric, curve saliency, for motion curves that specifies the important frames of the motion. In the second step, we detect the final key frames by clustering the computed important frames. As a result of our experimental results, on t...