Exploring visual features through Gabor representations for facial expression detection

Gabor representations have been widely used in facial analysis (face recognition, face detection and facial expression detection) due to their biological relevance and computational properties. Two popular Gabor representations used in literature are: 1) Log-Gabor and 2) Gabor energy filters. Even though these representations are somewhat similar, they also have distinct differences as the Log-Gabor filters mimic the simple cells in the visual cortex while the Gabor energy filters emulate the complex cells, which causes subtle differences in the responses. In this paper, we analyze the difference between these two Gabor representations and quantify these differences on the task of facial action unit (AU) detection. In our experiments conducted on the Cohn-Kanade dataset, we report an average area underneath the ROC curve (A`) of 92.60% across 17 AUs for the Gabor energy filters, while the Log-Gabor representation achieved an average A` of 96.11%. This result suggests that small spatial differences that the Log-Gabor filters pick up on are more useful for AU detection than the differences in contours and edges that the Gabor energy filters extract.

Cascading appearance-based features for visual voice activity detection

The detection of voice activity is a challenging problem, especially when the level of acoustic noise is high. Most current approaches only utilise the audio signal, making them susceptible to acoustic noise. An obvious approach to overcome this is to use the visual modality. The current state-of-the-art visual feature extraction technique is one that uses a cascade of visual features (i.e. 2D-DCT, feature mean normalisation, interstep LDA). In this paper, we investigate the effectiveness of this technique for the task of visual voice activity detection (VAD), and analyse each stage of the cascade and quantify the relative improvement in performance gained by each successive stage. The experiments were conducted on the CUAVE database and our results highlight that the dynamics of the visual modality can be used to good effect to improve visual voice activity detection performance.

Visual voice activity detection using frontal versus profile views

Visual activity detection of lip movements can be used to overcome the poor performance of voice activity detection based solely in the audio domain, particularly in noisy acoustic conditions. However, most of the research conducted in visual voice activity detection (VVAD) has neglected addressing variabilities in the visual domain such as viewpoint variation. In this paper we investigate the effectiveness of the visual information from the speaker’s frontal and profile views (i.e left and right side views) for the task of VVAD. As far as we are aware, our work constitutes the first real attempt to study this problem. We describe our visual front end approach and the Gaussian mixture model (GMM) based VVAD framework, and report the experimental results using the freely available CUAVE database. The experimental results show that VVAD is indeed possible from profile views and we give a quantitative comparison of VVAD based on frontal and profile views The results presented are useful in the development of multi-modal Human Machine Interaction (HMI) using a single camera, where the speaker’s face may not always be frontal.

Global flash multifocal electroretinogram : early detection of local functional changes and its correlations with optical coherence tomography and visual field loss in diabetic eyes

Purpose: To investigate the correlations of the global flash multifocal electroretinogram (MOFO mfERG) with common clinical visual assessments – Humphrey perimetry and Stratus circumpapillary retinal nerve fiber layer (RNFL) thickness measurement in type II diabetic patients. Methods: Forty-two diabetic patients participated in the study: ten were free from diabetic retinopathy (DR) while the remainder suffered from mild to moderate non-proliferative diabetic retinopathy (NPDR). Fourteen age-matched controls were recruited for comparison. MOFO mfERG measurements were made under high and low contrast conditions. Humphrey central 30-2 perimetry and Stratus OCT circumpapillary RNFL thickness measurements were also performed. Correlations between local values of implicit time and amplitude of the mfERG components (direct component (DC) and induced component (IC)), and perimetric sensitivity and RNFL thickness were evaluated by mapping the localized responses for the three subject groups. Results: MOFO mfERG was superior to perimetry and RNFL assessments in showing differences between the diabetic groups (with and without DR) and the controls. All the MOFO mfERG amplitudes (except IC amplitude at high contrast) correlated better with perimetry findings (Pearson’s r ranged from 0.23 to 0.36, p<0.01) than did the mfERG implicit time at both high and low contrasts across all subject groups. No consistent correlation was found between the mfERG and RNFL assessments for any group or contrast conditions. The responses of the local MOFO mfERG correlated with local perimetric sensitivity but not with RNFL thickness. Conclusion: Early functional changes in the diabetic retina seem to occur before morphological changes in the RNFL.

Enabling aircraft emergency landings using active visual site detection

The ability to automate forced landings in an emergency such as engine failure is an essential ability to improve the safety of Unmanned Aerial Vehicles operating in General Aviation airspace. By using active vision to detect safe landing zones below the aircraft, the reliability and safety of such systems is vastly improved by gathering up-to-the-minute information about the ground environment. This paper presents the Site Detection System, a methodology utilising a downward facing camera to analyse the ground environment in both 2D and 3D, detect safe landing sites and characterise them according to size, shape, slope and nearby obstacles. A methodology is presented showing the fusion of landing site detection from 2D imagery with a coarse Digital Elevation Map and dense 3D reconstructions using INS-aided Structure-from-Motion to improve accuracy. Results are presented from an experimental flight showing the precision/recall of landing sites in comparison to a hand-classified ground truth, and improved performance with the integration of 3D analysis from visual Structure-from-Motion.

Novelty-based visual obstacle detection in agriculture

This paper describes a novel obstacle detection system for autonomous robots in agricultural field environments that uses a novelty detector to inform stereo matching. Stereo vision alone erroneously detects obstacles in environments with ambiguous appearance and ground plane such as in broad-acre crop fields with harvested crop residue. The novelty detector estimates the probability density in image descriptor space and incorporates image-space positional understanding to identify potential regions for obstacle detection using dense stereo matching. The results demonstrate that the system is able to detect obstacles typical to a farm at day and night. This system was successfully used as the sole means of obstacle detection for an autonomous robot performing a long term two hour coverage task travelling 8.5 km.

Complete-linkage clustering for voice activity detection in audio and visual speech

We propose a novel technique for conducting robust voice activity detection (VAD) in high-noise recordings. We use Gaussian mixture modeling (GMM) to train two generic models; speech and non-speech. We then score smaller segments of a given (unseen) recording against each of these GMMs to obtain two respective likelihood scores for each segment. These scores are used to compute a dissimilarity measure between pairs of segments and to carry out complete-linkage clustering of the segments into speech and non-speech clusters. We compare the accuracy of our method against state-of-the-art and standardised VAD techniques to demonstrate an absolute improvement of 15% in half-total error rate (HTER) over the best performing baseline system and across the QUT-NOISE-TIMIT database. We then apply our approach to the Audio-Visual Database of American English (AVDBAE) to demonstrate the performance of our algorithm in using visual, audio-visual or a proposed fusion of these features.

Acoustic adaptation in cross database audio visual SHMM training for phonetic spoken term detection

Visual information in the form of lip movements of the speaker has been shown to improve the performance of speech recognition and search applications. In our previous work, we proposed cross database training of synchronous hidden Markov models (SHMMs) to make use of external large and publicly available audio databases in addition to the relatively small given audio visual database. In this work, the cross database training approach is improved by performing an additional audio adaptation step, which enables audio visual SHMMs to benefit from audio observations of the external audio models before adding visual modality to them. The proposed approach outperforms the baseline cross database training approach in clean and noisy environments in terms of phone recognition accuracy as well as spoken term detection (STD) accuracy.

Incorporating visual information for spoken term detection

Spoken term detection (STD) is the task of looking up a spoken term in a large volume of speech segments. In order to provide fast search, speech segments are first indexed into an intermediate representation using speech recognition engines which provide multiple hypotheses for each speech segment. Approximate matching techniques are usually applied at the search stage to compensate the poor performance of automatic speech recognition engines during indexing. Recently, using visual information in addition to audio information has been shown to improve phone recognition performance, particularly in noisy environments. In this paper, we will make use of visual information in the form of lip movements of the speaker in indexing stage and will investigate its effect on STD performance. Particularly, we will investigate if gains in phone recognition accuracy will carry through the approximate matching stage to provide similar gains in the final audio-visual STD system over a traditional audio only approach. We will also investigate the effect of using visual information on STD performance in different noise environments.

A phenanthrene based highly selective fluorogenic and visual sensor for Cu2+ ion with nanomolar detection limit and its application in live cell imaging

A new phenanthrene based chemosensor has been synthesized and investigated to act as highly selective fluorescence and visual sensor for Cu2+ ion with very low detection limit of 1.58 nM: this has also been used to image Cu2+ in human cervical HeLa cancer cells. (C) 2012 Elsevier B.V. All rights reserved.

Automated Detection of Concrete Columns from Visual Data

After earthquakes, licensed inspectors use the established codes to assess the impact of damage on structural elements. It always takes them days to weeks. However, emergency responders (e.g. firefighters) must act within hours of a disaster event to enter damaged structures to save lives, and therefore cannot wait till an official assessment completes. This is a risk that firefighters have to take. Although Search and Rescue Organizations offer training seminars to familiarize firefighters with structural damage assessment, its effectiveness is hard to guarantee when firefighters perform life rescue and damage assessment operations together. Also, the training is not available to every firefighter. The authors therefore proposed a novel framework that can provide firefighters with a quick but crude assessment of damaged buildings through evaluating the visible damage on their critical structural elements (i.e. concrete columns in the study). This paper presents the first step of the framework. It aims to automate the detection of concrete columns from visual data. To achieve this, the typical shape of columns (long vertical lines) is recognized using edge detection and the Hough transform. The bounding rectangle for each pair of long vertical lines is then formed. When the resulting rectangle resembles a column and the material contained in the region of two long vertical lines is recognized as concrete, the region is marked as a concrete column surface. Real video/image data are used to test the method. The preliminary results indicate that concrete columns can be detected when they are not distant and have at least one surface visible.

Comparison between visual and passive acoustic detection of finless porpoises in the Yangtze river, China

Recently, sonar signals and other sounds produced by cetaceans have been used for acoustic detection of individuals and groups in the wild. However, the detection probability ascertained by concomitant visual survey has not been demonstrated extensively. The finless porpoises (Neophocaena phocaenoides) have narrow band and high-frequency sonar signals, which are distinctive from background noises. Underwater sound monitoring with hydrophones (B&K8103) placed along the sides of a research vessel, concurrent with visual observations was conducted in the Yangtze River from Wuhan to Poyang Lake in 1998 in China. The peak to peak detection threshold was set at 133 dB re 1 mu Pa. With this threshold level, porpoises could be detected reliably within 300 m of the hydrophone. In a total of 774-km cruise, 588 finless porpoises were sighted by visual observation and 44 864 ultrasonic pulses were recorded by the acoustical observation system. The acoustic monitoring system could detect the presence of the finless porpoises 82% of the time. A false alarm in the system occurred with a frequency of 0.9%. The high-frequency acoustical observation is suggested as an effective method for field surveys of small cetaceans, which produce high-frequency sonar signals. (C) 2001 Acoustical Society of America.