The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Compressive Sensing (CS) is a popular signal processing technique, that can exactly reconstruct a signal given a small number of random projections of the original signal, provided that the signal is sufficiently sparse. We demonstrate the applicability of CS in the field of gait recognition as a very effective dimensionality reduction technique, using the gait energy image (GEI) as the feature extraction...
This paper presents a method for recovering the reflectance and shape of a surface from a single hyper spectral image. To do this, we depart from a reflectance model based on a physical interpretation of the reflection process where the geometry of the scene and the properties of the object surface describe the image formation process. With the model in hand, we propose a solution to the recovery...
Tracking pedestrians through a network of cameras is a difficult and nuanced problem that is of significant interest in visual surveillance. This paper describes an online system for tracking multiple people as they move around a small camera network with partially-overlapping fields of view. Multi-object tracking is performed on each camera feed resulting in single-camera pedestrian tracks. Multi-camera...
Due to the complex nature of the driving environment, obstacle tracking systems are required to rely on intermediate dynamic information, before the obstacle is fully reconstructed. This paper presents an obstacle estimation system which uses the advantages of a particle-based occupancy grid tracking solution. The initial measurement data is a raw occupancy map extracted from dense stereovision-derived...
We propose a real-time multi-camera tracking approach to follow vehicles in a tunnel surveillance environment with multiple non-overlapping cameras. In such system, vehicles have to be tracked in each camera and passed correctly from one camera to another through the tunnel. This task becomes extremely difficult when intra-camera errors are accumulated. Most typical issues to solve in tunnel scenes...
Measuring the similarity between articulated shapes is a fundamental yet challenging problem. This paper proposes a novel shape descriptor based on Width Distributions (WD), which is robust to articulations. We show that the width distributions are articulation insensitive yet descriptive to distinguish different shapes with varied part structures. With measurements on distributions only, the proposed...
For image classification applications it is often useful to generate a compact representation of the texture of an image region. The conventional representation of image textures using extracted Gabor wavelet coefficients often yields poor performance when classifying scaled and rotated versions of image regions. In this paper we propose a scale and rotation invariant feature generation procedure...
In this paper, we propose novel algorithmic models based on information fusion and feature transformation in cross-modal subspace for different types of residue features extracted from several intra-frame and inter-frame pixel sub-blocks in video sequences for detecting digital video tampering or forgery. An evaluation of proposed residue features -- the noise residue features and the quantization...
Image matting refers to the problem of accurately extracting foreground objects in images and video. The most recent work by Levin, Lischinski and Weiss (2008) in natural image matting relies on the local smoothness assumptions on foreground and background colors on which a cost function is established. The closed-form solution has been derived based on certain degree of user inputs. In this paper,...
Facial expression recognition (FER) algorithms mainly focus on classification into a small discrete set of emotions or representation of emotions using facial action units (AUs). Dimensional representation of emotions as continuous values in an arousal-valence space is relatively less investigated. It is not fully known whether fusion of geometric and texture features will result in better dimensional...
Minimal projective ghost functions make good watermark labels for embedding into images. Although fragile to hacking attacks, they are near-to-invisible because of their distributed, random appearance and their binary, zero-mean statistics. They have the strong correlation properties needed to extract a low-intensity watermark from bright image data. We present a method to embed, concurrently, up...
In this paper, two energy minimization based image segmentation methods are investigated. The formulations of both methods-level set and Otsu thresholding-are detailed. Several experiments are carried out and the comparison is conducted in terms of segmentation results, CPU time cost and adaptability to multi-phase images. The pros and cons of the methods are also discussed. As a by-product, we propose...
This paper presents an efficient method in deciding macro-block mode and selecting prediction mode for intra prediction in H.264/AVC high profile. H.264/AVC supports nine intra prediction modes for luminance 4x4 and 8x8 blocks. The predictors include 8 directional modes and intra DC mode, which is non-directional mode. For luminance 16x16 blocks, the following 4 intra prediction modes are used: Vertical,...
Nowadays leaf image classification is very useful for both botanists and ordinary users since advanced imaging devices such as smart phones make it ever easier to capture leaf images for various tasks such as retrieval and classification. Most of existing approaches mainly utilize global shape features. In this paper, we propose to improve leaf image classification by taking both global features and...
This paper presents a fast intra mode decision algorithm using the sum of absolute transformed differences (SATD) in order to reduce the encoding time. Our algorithm consists of two main techniques: Early block-mode decision and candidate directional-mode selection. First, we use the early block-mode decision method, I4MB skip, to avoid unnecessary computation for I4MB. If all the 16 4x4sub-blocks...
Measuring coral growth rate is essential for monitoring coral reef health, and part of the process involves directional analysis of x-ray images of coral sections. The monogenic signal is useful for this application as it represents an image in terms of intrinsically-1D feature type (phase), strength (amplitude) and orientation. At certain locations the monogenic signal may give orientation errors,...
When a single-sensor digital camera is used for full colour image capture, only one colour for each pixel is acquired through the colour filter array which is placed in front of the image sensor. CFA demosaicking is a process for the estimation of the other missing colour values in order to produce a full colour image. To preserve sharp colour edges and avoid colour artifacts, it is desirable not...
Biometric systems play an important role in the field of information security as they are extremely required for user authentication. Automatic signature recognition and verification is one of the biometric techniques, which is currently receiving renewed interest and is only one of several techniques used to verify the identities of individuals. Signatures provide a secure means for confirmation...
This paper presents a method of capturing a document with a single camera based on the Hough transform with reduced angle domain. Document image capturing with a camera has some advantages over traditional flatbed scanners. A problem of capturing a document with a camera is to find a paper region in an image and find robust features to remove perspective distortions. If we have boundary information...
Many fast block-matching algorithms (BMAs) reduce computational complexity by screening the number of checking points. Although many fast BMAs reduce computations, sometimes they should endure matching errors in comparison with full-search algorithm (FSA). In this paper, a novel fast BMA for constrained one-bit transform (C1BT)-based motion estimation is proposed in order to decrease the calculations...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.