Abstracts:Subspace analysis or dimensionality reduction techniques are becoming very popular for many computer vision tasks such as image recognition. Most of such techniques deal with optimizing a cost function based on some criteria imposed on either projections of data or on the basis of projection space. NPP and ONPP are such linear methods that preserve local linear relationship within the neighborhood, with two different constraints, normalized projection and orthogonal basis of subspace respectively. This article proposes a method, ONPPn, that finds a subspace which satisfies two constraints namely, normalization and orthogonality. The article also provides two-dimensional variant of ONPPn. Experiments show that ONPPn outperforms its NPP and ONPP versions in image recognition tasks, whereas 2D-ONPPn outperforms 2D-ONPP by huge margin but does not perform as good as 2D-NPP. 2D-NPP as well as 2D-ONPP are not suitable for reconstruction task, but the proposed method 2D-ONPPn overcomes drawbacks of existing methods and is best suited for image reconstruction, too.

Abstracts:Combating against camcorder piracy requires identification of the theater and show time information, followed by the estimation of camcorder location in a theater from which an illegal recording was made, in order to find out the pirate and limit the number of pirate suspects. State-of-the-art pirate position estimation frameworks employ watermarking techniques to approximate the position of the pirate in a movie theater. However, watermarks are fragile in nature and involve complex procedures, which may damage the video content. To solve this, a novel forensic tracking framework, which exploits visual-audio fingerprints for estimating the location of the pirate in a theater without embedding digital watermarks is presented. Precisely, the proposed framework first spatio-temporally aligns the source movie and captured video contents, then estimates the geometric distortions and consequently derives the illegal capture location in a theater by exploiting multimodal features. The case study results in the form of sophisticated In-theater experiments prove that, it is certainly possible to estimate the illegal capture location in a theater with a mean absolute error of (38.25, 22.45, 11.11) cm, by employing multimodal fingerprints. In this way, the proposed article demonstrates a brand-new application of video fingerprinting for investigating the location of illegal camcorder capture in a theater, which is applicable for digital cinema applications.

Abstracts:In the last decade, research in biometrics has been focused on augmenting the algorithmic performance to address a growing range of applications, not limited to person authentication/recognition. The concept of context awareness emerged as a possible key-factor for both performance optimization and operational adaptation of the capture, extraction, matching and decision stages. This may be particularly effective for multi-biometrics systems. The knowledge of the context in which a task is being performed, may provide useful information to the system in several manners. For example, it may allow to adapt to a specific environmental condition, such as shadow or light exposure. On the other hand, it may be possible to select the best available algorithm, among a given set to address the task at hand, which best performs within the given context. This paper aims to provide an overall vision of the main contributions available so far in the field of context-aware biometric systems and methods. The survey is not confined to a particular biometric modality or processing stage, but rather spans the state of the art of several biometric modalities and approaches. A taxonomy of context-aware biometric systems and methods is also proposed, along with a comparison of their features, aims and performances. The analysis will be complemented with a critical discussion about the state of the art also suggesting some future application scenarios.

Abstracts:Squared planar markers have become a popular method for pose estimation in applications such as autonomous robots, unmanned vehicles and virtual trainers. The markers allow estimating the position of a monocular camera with minimal cost, high robustness, and speed. One only needs to create markers with a regular printer, place them in the desired environment so as to cover the working area, and then registering their location from a set of images.

Abstracts:Gait recognition systems allow identification of users relying on features acquired from their body movement while walking. This paper discusses the main factors affecting the gait features that can be acquired from a 2D video sequence, proposing a taxonomy to classify them across four dimensions. It also explores the possibility of obtaining users' gait features from the shadow silhouettes by proposing a novel gait recognition system. The system includes novel methods for: (i) shadow segmentation, (ii) walking direction identification, and (iii) shadow silhouette rectification.

Abstracts:We propose a novel concept of asymmetric feature maps (AFM), which allows to evaluate multiple kernels between a query and database entries without increasing the memory requirements. To demonstrate the advantages of the AFM method, we derive an efficient contour match kernel – short vector image representation that, due to asymmetric feature maps, supports efficient scale and translation invariant sketch-based image retrieval. Unlike most of the short-code based retrieval systems, the proposed method provides the query localization in the retrieved image. The efficiency of the search is boosted by approximating a 2D translation search via trigonometric polynomial of scores by 1D projections. The projections are a special case of AFM. An order of magnitude speed-up is achieved compared to traditional trigonometric polynomials. The results are boosted by an image-based average query expansion approach and, without any learning, significantly outperform the state-of-the-art hand-crafted descriptors on standard benchmarks. Our method competes well with recent CNN-based approaches that require large amounts of labeled sketches, images and sketch-image pairs.