Additional Camera Research Articles

Full-Spectrum Endoscopy Improves Surveillance for Dysplasia in Patients With Inflammatory Bowel Diseases

Inflammatory bowel diseases (IBDs) increase the risk of colorectal cancer. Surveillance colonoscopy with chromoendoscopy is recommended, but conventional forward-viewing colonoscopy (FVC) detects dysplasia with lowlevels of sensitivity. Full-spectrum endoscopy (FUSE) incorporates 2 additional lateral cameras to the forward camera of the colonoscope, allowing endoscopists to view behind folds and in blind spots, which might increase dysplasia detection. We compared FUSE vs FVC in the detection of dysplasia in patients with IBDs. We performed a prospective, randomized, cross-over, tandem colonoscopy study comparing FVC vs FUSE in 52 subjects with IBD undergoing surveillance for neoplasia in Australia (23 with Crohn's colitis, 29 with ulcerative colitis; median age, 45.0 y; 60% male; mean IBD duration, 16.4 y). All subjects met national IBD surveillance inclusion criteria; 27 were assigned randomly to groups that underwent FVC followed by FUSE, and 25 were assigned to groups that underwent FUSE followed by FVC. All procedures were performed from February 2014 through December 2015. Random biopsy specimens were collected and visible lesions were collected; all were analyzed histologically. The primary end point was dysplasia missed by the first colonoscopy detected by the second colonoscopy. Dysplasia was diagnosed by an expert gastrointestinal pathologist blinded to the colonoscope allocation in consensus with a second expert pathologist. FVC missed 71.4% of dysplastic lesions per lesion whereas FUSE missed 25.0% per lesion (P= .0001); FVC missed 75.0% of dysplastic lesions per subject and FUSE missed 25.0% per subject (P= .046). FUSE identified a mean of 0.37 dysplastic lesions and FVC identified a mean of 0.13 dysplastic lesions (P= .044). The total colonoscopy times were similar (21.2 min for FUSE vs 19.1 min for FVC; P= .32), but withdrawal time was significantly longer for FUSE (15.8 min) than for FVC (12.0 min) (P= .03). Correcting for per-unit withdrawal time, the mean dysplasia miss rate per subject was significantly lower for FUSE (0.19) than for FVC (0.83; P<.0001). Targeted tissue acquisition identified significantly more dysplastic lesions than random biopsies (P < .0001). In a prospective cross-over study of IBD patients undergoing surveillance colonoscopy, we found panoramic views obtained by full-spectrum endoscopy increased thenumber of dysplastic lesions detected, compared with conventional forward-viewing colonoscopy. Trial no: ACTRN12616000047493.

Flexible Gastro-intestinal Endoscopy — Clinical Challenges and Technical Achievements

Flexible gastro-intestinal (GI) endoscopy is an integral diagnostic and therapeutic tool in clinical gastroenterology. High quality standards for safety, patients' comfort, and efficiency have already been achieved. Clinical challenges and technical approaches are discussed in this short review.Image enhanced endoscopy for further characterization of mucosal and vascular patterns includes dye-spray or virtual chromoendoscopy. For confocal laser endoscopy, endocytoscopy, and autofluorescence clinical value has not yet been finally evaluated. An extended viewing field provided by additional cameras in new endoscopes can augment detection of polyps behind folds. Attachable caps, flaps, or balloons can be used to flatten colonic folds for better visualization and stable position.Variable stiffness endoscopes, radiation-free visualization of endoscope position, and different overtube devices help reducing painful loop formation in clinical routine. Computer assisted and super flexible self-propelled colonoscopes for painless sedation-free endoscopy need further research. Single-use devices might minimize the risk of infection transmission in the future.Various exchangeable accessories are available for resection, dissection, tunneling, hemostasis, treatment of stenosis and closure of defects, including dedicated suturing devices. Multiple arm flexible devices controlled via robotic platforms for complex intraluminal and transmural endoscopic procedures require further improvement.

HAZE AT OCCATOR CRATER ON DWARF PLANET CERES

A diurnal varying haze layer at the bright spots of Occator on dwarf planet Ceres has been reported from images of the Dawn Framing Camera. This finding is supported by ground-based observations revealing diurnal albedo changes at Occator's longitude. In the present work, we further investigate the previously reported haze phenomenon in more detail using additional Framing Camera images. We demonstrate that the light scattering behavior at the central floor of Occator is different compared to a typical cerean surface and is likely inconsistent with a pure solid surface scatterer. The identified deviation is best explained by an additional component to the scattered light of the surface, i.e., a haze layer. Our results support the water vapor detection by Herschel observations though the existence of a tenuous cerean exosphere is not yet confirmed.

Expanding the Limits of Vision‐based Localization for Long‐term Route‐following Autonomy

Vision‐based, autonomous, route‐following algorithms enable robots to autonomously repeat manually driven routes over long distances. Through the use of inexpensive, commercial vision sensors, these algorithms have the potential to enable robotic applications across multiple industries. However, in order to extend these algorithms to long‐term autonomy, they must be able to operate over long periods of time. This poses a difficult challenge for vision‐based systems in unstructured and outdoor environments, where appearance is highly variable. While many techniques have been developed to perform localization across extreme appearance change, most are not suitable or untested for vision‐in‐the‐loop systems such as autonomous route following, which requires continuous metric localization to keep the robot driving. In this paper, we present a vision‐based, autonomous, route‐following algorithm that combines multiple channels of information during localization to increase robustness against daily appearance change such as lighting. We explore this multichannel visual teach and repeat framework by adding the following channels of information to the basic single‐camera, gray‐scale, localization pipeline: images that are resistant to lighting change and images from additional stereo cameras to increase the algorithm's field of view. Using these methods, we demonstrate robustness against appearance change through extensive field deployments spanning over 26 km with an autonomy rate greater than 99.9%. We furthermore discuss the limits of this system when subjected to harsh environmental conditions by investigating keypoint match degradation through time.

Blur robust optical flow using motion channel

It is hard to estimate optical flow given a realworld video sequence with camera shake and other motion blur. In this paper, we first investigate the blur parameterisation for video footage using near linear motion elements. We then combine a commercial 3D pose sensor with an RGB camera, in order to film video footage of interest together with the camera motion. We illustrate that this additional camera motion/trajectory channel can be embedded into a hybrid framework by interleaving an iterative blind deconvolution and warping based optical flow scheme. Our method yields improved accuracy within three other state-of-the-art baselines given our proposed ground truth blurry sequences; and several other realworld sequences filmed by our imaging system.

3D Model Reconstruction System Development Based on Laser-Vision Technology

We propose a three-dimensional (3D) scanning system based on laser-vision technique and rotary mechanism combination for automatic 3D model reconstruction. The proposed scanning system consists of a laser projector, camera, and turntable. For laser-camera calibration, a new and simple method is applied. The 3D point cloud data of the surface of the scanned object are fully collected by integrating the extracted laser profiles from the laser stripe images corresponding to the rotary angles of the rotary mechanism. The obscured laser profile problem is solved by an additional camera at another viewpoint. From the collected 3D point cloud data, the 3D model of the scanned object is reconstructed based on the facet-representation method. The reconstructed 3D models showed the effectiveness and applicability of the proposed 3D scanning system in many 3D model-based applications.

Single-shot 3D flame diagnostic based on volumetric laser induced fluorescence (VLIF)

We demonstrate single-shot three-dimensional (3D) flame measurements based on volumetric laser-induced fluorescence (VLIF). For this effort, the excitation laser beam was expanded into a “slab” with a height of 40mm and thickness of 10mm to excite the CH radicals in the target flames. The volumetric LIF signals were then simultaneously captured by five intensified cameras, and the images were fed into a tomographic algorithm as inputs to obtain the 3D distribution of CH. The results show that single-shot 3D measurements with sufficient accuracy can be obtained in a volume of 9.3×9.3×32.7mm3 with an excitation pulse energy of ∼10mJ. With a total of five intensified cameras, 3D measurements were obtained with a voxel size of 0.15mm over a total of 64×64×224≈106 voxels, which can be further improved by the use of additional cameras or advanced tomographic inversion techniques. Comparison with PLIF measurements suggested that the spatial resolution of the CH-based VLIF technique to characterize the flame surface was about 0.4mm. Measurements were first performed on stable laminar flames, for comparison and validation against PLIF, and then on turbulent jet flames, to illustrate the feasibility and accuracy of VLIF for instantaneous 3D flame measurements and motivate further investigation.

Optical Mapping of Membrane Potential and Epicardial Deformation in Beating Hearts

Cardiac optical mapping uses potentiometric fluorescent dyes to image membrane potential (Vm). An important limitation of conventional optical mapping is that contraction is usually arrested pharmacologically to prevent motion artifacts from obscuring Vm signals. However, these agents may alter electrophysiology, and by abolishing contraction, also prevent optical mapping from being used to study coupling between electrical and mechanical function. Here, we present a method to simultaneously map Vm and epicardial contraction in the beating heart. Isolated perfused swine hearts were stained with di-4-ANEPPS and fiducial markers were glued to the epicardium for motion tracking. The heart was imaged at 750 Hz with a video camera. Fluorescence was excited with cyan or blue LEDs on alternating camera frames, thus providing a 375-Hz effective sampling rate. Marker tracking enabled the pixel(s) imaging any epicardial site within the marked region to be identified in each camera frame. Cyan- and blue-elicited fluorescence have different sensitivities to Vm, but other signal features, primarily motion artifacts, are common. Thus, taking the ratio of fluorescence emitted by a motion-tracked epicardial site in adjacent frames removes artifacts, leaving Vm (excitation ratiometry). Reconstructed Vm signals were validated by comparison to monophasic action potentials and to conventional optical mapping signals. Binocular imaging with additional video cameras enabled marker motion to be tracked in three dimensions. From these data, epicardial deformation during the cardiac cycle was quantified by computing finite strain fields. We show that the method can simultaneously map Vm and strain in a left-sided working heart preparation and can image changes in both electrical and mechanical function 5 min after the induction of regional ischemia. By allowing high-resolution optical mapping in the absence of electromechanical uncoupling agents, the method relieves a long-standing limitation of optical mapping and has potential to enhance new studies in coupled cardiac electromechanics.

Fuse and image enhancement.

Colorectal cancer (CRC) is the most devastating consequence of long-standing colitis in inflammatory bowel disease (IBD) patients. In a population-based study, the cumulative prevalence rates of CRC were 1%, 3%, and 7% after 10, 20, and 30 years of disease activity, respectively.1 The higher than expected rate of CRC underlies the importance of dysplasia surveillance programs in this population. However, one of the challenges faced by endoscopists is that the dysplasia seen in this cohort is mostly flat and subtle, which can be easily missed in the midst of mucosal inflammation and pseudopolyps. In addition, the use of low-definition, standard white-light colonoscopy and traditional practice of random quadrantic biopsy that only sample < 0.05% of the entire colonic mucosa lacks acceptable sensitivity and specificity in dysplasia identification and is associated with high rate of interval colitis-associated cancers.2 There is compelling evidence to support chromoendoscopy (CE) in dysplasia detection in long-standing IBD patients, and this technique has been endorsed for surveillance colonoscopy by all major guidelines including American Gastroenterological Association, American College of Gastroenterology, Crohn's and Colitis Foundation of America, European Crohn's and Colitis Organization, British Society of Gastroenterology, American Society for Gastrointestinal Endoscopy, the Australian National Health and Medical Research Council, and the Surveillance for Colorectal Endoscopic Neoplasia Detection and Management in Inflammatory Bowel Disease Patients: International Consensus Recommendations. However, despite guideline recommendations, CE has not been widely adopted into clinical practice, mainly related to the cumbersome and time-consuming technique.3 Many endoscopists prefer to use non-dye-based digital image enhancement, such as narrow band imaging, although there is currently a lack of evidence that narrow band imaging improves the detection of dysplasia over and above CE in IBD patients.4 Recent innovations, such as confocal laser endomicroscopy and autofluorescence imaging, have been evaluated in dysplasia surveillance in IBD and appear promising in improving neoplasia detection.5-7 However, studies performed to date are small, and larger studies are clearly needed to evaluate the accuracy of these new technologies to determine their role in dysplasia screening algorithm for IBD patients. Other new imaging modalities in development are Fuji Intelligent Chromoendoscopy and i-scan that utilize surface and tone enhancement via a digital image processing algorithm, which avoids the use of contrast to better define and outline mucosal structures in real time. However, there is no study to date on the use of such technology in dysplasia surveillance in IBD patients.8 Optical coherence tomography is another new innovative imaging modality that has mainly been evaluated in Barrett's surveillance and not yet in dysplasia detection in IBD. This technology utilizes an optical probe that is balloon-centered that can be placed endoscopically to provide radial imaging scan using infrared light, capturing 1200 images over 90 s. However, given the limited performance data currently available, this technology is mainly assessed in research setting.9 Recently, a new imaging modality became available in Australia known as full spectrum endoscopy (FUSE) that maintains the capability of an adult colonoscope but has additional camera lenses to the right and left sides of the tip of the scope providing a near-panoramic 330-degree endoscopic view of the mucosa as opposed to the restricted 170-degree endoscopic view from conventional forward viewing colonoscopes. Improved visualization of the side walls, blind spots, and behind mucosal folds allows more colonic lesions to be identified, hence reducing polyp miss rates. A recent tandem back-to-back screening colonoscopy study of a non-IBD population demonstrated a significant reduction in sporadic colonic adenoma miss rates from 41% to 7% using forward viewing colonoscopes versus FUSE, respectively.10 FUSE colonoscopy has yet to be evaluated for IBD dysplasia and the benefit of FUSE in comparison against the current gold standard technique of dye-spray CE also remains unknown. In summary, improvements in the optics of endoscopes and the use of CE have improved our ability to detect dysplasia. However, the current strategies used still remains inadequate, and the development of CRC continues to fuel ongoing research in advanced endoscopic imaging techniques to improve the efficacy of dysplasia surveillance. These emerging modalities appear promising and may change our traditional approach to dysplasia screening in IBD patients in the years ahead. A major hurdle continues to be the low uptake of CE in IBD surveillance.

Optimizing Detection Rate and Characterization of Subtle Paroxysmal Neonatal Abnormal Facial Movements with Multi-Camera Video-Electroencephalogram Recordings.

Objectives We retrospectively analyze the diagnostic accuracy for paroxysmal abnormal facial movements, comparing one camera versus multi-camera approach. Background Polygraphic video-electroencephalogram (vEEG) recording is the current gold standard for brain monitoring in high-risk newborns, especially when neonatal seizures are suspected. One camera synchronized with the EEG is commonly used. Methods Since mid-June 2012, we have started using multiple cameras, one of which point toward newborns' faces. We evaluated vEEGs recorded in newborns in the study period between mid-June 2012 and the end of September 2014 and compared, for each recording, the diagnostic accuracies obtained with one-camera and multi-camera approaches. Results We recorded 147 vEEGs from 87 newborns and found 73 episodes of paroxysmal facial abnormal movements in 18 vEEGs of 11 newborns with the multi-camera approach. By using the single-camera approach, only 28.8% of these events were identified (21/73). Ten positive vEEGs with multicamera with 52 paroxysmal facial abnormal movements (52/73, 71.2%) would have been considered as negative with the single-camera approach. Conclusions The use of one additional facial camera can significantly increase the diagnostic accuracy of vEEGs in the detection of paroxysmal abnormal facial movements in the newborns.

AliveCor Mobile ECG: A smartphone with heart

S have impacted our lives by providing what many consider important necessities and utilities in the form of a compact, portable handheld device. High-tech accessories such as smart watches, Bluetooth headsets, and additional camera lenses further enhance the capabilities of our smartphones, making some of us feel inseparable from these devices. Now, the scope of functions made possible by smartphone accessories includes health care services. For example, AliveCor Mobile ECG is a new electrocardiogram (ECG) monitor that can be attached to the back of an iPhone or Android smartphone to help cardiac patients track the status of their heart.

Flexible calibration method for an inner surface detector based on circle structured light.

A new calibration method for an inner surface detector based on circle structured light is proposed in this study. Compared with existing methods, this technique is more flexible and practical and only requires a blank planar board and an additional camera, which is precalibrated under the detector's coordinate system. The board is observed by the detector and the additional camera at a few (at least two) different orientations, which need not be known. The mathematical model of this method considers different alignment errors, which are disregarded in existing methods; therefore, precise assembly is not required. The binocular intersection algorithm is used to calculate the coordinates of the calibration points. The measurement system calibrated by this method performs well in the field test in which the maximum relative error of the measured values is less than 0.18%. The experimental result indicates that this method is highly accurate and can be easily applied in inner surface detection.

Safety performance testing of climbing helmet-mounted cameras

BackgroundNo research establishing the effects of climbing helmet-mounted cameras on head injury biomechanics.ObjectiveEstablish the potential effects of climbing helmet-mounted cameras on the injury risks associated with falling object strikes and...

Soft-biometrics evaluation for people re-identification in uncontrolled multi-camera environments

A novel method for person identification based on soft-biometrics and oriented to work in real video surveillance environments is proposed in this paper. Thus, an evaluation of relevance’s level of several appearance features is carried out with this purpose. First, a bag-of-soft-biometric features related to color, texture, local features, and geometry are extracted from individuals. The relevance of each feature has been deeply analyzed through different proposed methods. Features are ranked and weighted according to their relevance value. Later, each method is evaluated under two different scenarios: mono-camera and multi-camera surveillance images. In order to test the system in a realistic way, it has been evaluated over standard databases in the surveillance community: PETS 2006, PETS 2009, CAVIAR, SAIVT-SoftBio, and CAVIAR4REID. Moreover, a new database was acquired at Adolfo Suarez Madrid-Barajas international airport. This database was acquired under regular conditions and infrastructure of the Barajas airport, no additional camera or special settings were installed for this purpose. An analysis of relevance for each feature acquired in these two scenarios is presented. The results obtained demonstrate the promising potential of the soft-biometric approach. Finally, an optimal system configuration according to each scenario is obtained.

The Use of Camera Traps for Monitoring the Population of Long-Tailed Gorals

The long-tailed goral (Nemorhaedus caudatus) is a rare montane ungulate species with a patchy distribution. In the Sikhote-Alin Reserve, gorals occupy the northern part of their range, concentrated primarily in a small coastal area (6.4km2) in Abrek Urochishe. Our pilot study tested the feasibility of individual photo-identification of gorals and population size estimation using the capture–recapture method. We used 10 camera traps spaced 0.6–2km apart on coastal slopes to monitor the gorals. Four additional cameras were placed at the Reserve boundaries, mainly for law enforcement purposes, such as documenting trespassers. Between June 1 and December 31, 2013, we collected nearly 3000 photographs of gorals, 500 photographs of other wildlife, and 12 images of illegal activities within the Reserve. The total sampling effort was 1870 camera days. Photo data showed that goral horns are reliable biometric identifiers, distinguishable by size, shape, pattern, and the number of rings. The proportion of individually identified gorals in our photos was 0.64 (SE=0.05). Most individuals (45) were marked (i.e., first detected on camera) in the fall; therefore, preliminary estimates of the goral population size were made between October 11 and December 20, 2013. A closure test confirmed that the population was, in fact, closed (z=−2.670, P=0.004). The best-fit closed population multiple recapture model for our data was the heterogeneity model Mh (programme CAPTURE), which assumes an unequal capture probability (χ2=112.19; d.f.=9; P=0.000). The average goral capture probability was 0.16, and the corresponding population size was estimated at 90 individuals (SE=6.91; 95% CI: 77–125 individuals). The average goral population density in a 3.5km2 effective sampled area (56% of the entire plot area) was 25 individuals/km2 (SE=5.62). Extrapolation to locations that lacked data suggests that Abrek Urochishe supports a goral population of 160 individuals. Our results demonstrate that camera trap data can be used for photographic capture–recapture sampling of goral populations. This approach may be more effective than traditional visual surveys of montane ungulates that tend to underestimate the population abundance. The use of camera traps will undoubtedly enhance goral monitoring efforts, aiding in the conservation of this rare species.

Toward Welding Robot With Human Knowledge: A Remotely-Controlled Approach

This paper presents a remotely controlled welding scheme that enables transformation of human welder knowledge into a welding robot. In particular, a 6-DOF UR-5 industrial robot arm is equipped with sensors to observe the welding process, including a compact 3D weld pool surface sensing system and an additional camera to provide direct view of the work-piece. Human welder operates a virtual welding torch, whose motion is tracked by a Leap sensor. To remotely operate the robot based on the motion information from the Leap sensor, a predictive control approach is proposed to accurately track the human motion by controlling the speed of the robot arm movement. Tracking experiments are conducted to track both simulated movement with varying speed and actual human hand movement. It is found that the proposed predictive controller is able to track human hand movement with satisfactory accuracy. A welding experiment has also been conducted to verify the effectiveness of the proposed remotely-controlled welding system. A foundation is thus established to realize teleoperation and help transfer human knowledge to the welding robot.

Aerial multi-camera systems: Accuracy and block triangulation issues

Oblique photography has reached its maturity and has now been adopted for several applications. The number and variety of multi-camera oblique platforms available on the market is continuously growing. So far, few attempts have been made to study the influence of the additional cameras on the behaviour of the image block and comprehensive revisions to existing flight patterns are yet to be formulated. This paper looks into the precision and accuracy of 3D points triangulated from diverse multi-camera oblique platforms. Its coverage is divided into simulated and real case studies. Within the simulations, different imaging platform parameters and flight patterns are varied, reflecting both current market offerings and common flight practices. Attention is paid to the aspect of completeness in terms of dense matching algorithms and 3D city modelling – the most promising application of such systems. The experimental part demonstrates the behaviour of two oblique imaging platforms in real-world conditions. A number of Ground Control Point (GCP) configurations are adopted in order to point out the sensitivity of tested imaging networks and arising block deformations. To stress the contribution of slanted views, all scenarios are compared against a scenario in which exclusively nadir images are used for evaluation.

Gaze and performance of an elite karate exponent: a case study

The choice of attacks by Karate exponents is not well understood but is of importance to coaching practices. This case study explores gaze tracking sequences prior to attacking and seeks possible associative actions performed by the karateka. One Malaysian Karate exponent with over 8 years experience in the National team volunteered to participate in a simulation match of 2 minutes while wearing the Mobile Eye glasses (ASL, USA). Gaze was recorded at 25 Hz and was synchronised with an additional camera (25 Hz) recording the performance from just outside the mat. Gaze fixations were manually coded along 6 different areas (head, above the head, chest, abdominal, arm and leg) of the opponent. Also performance tactics (attack or counter) and actions (lunges, kick, punch and blocks) by the participant were manually notated. Gaze fixation sequences were selected from the start of an action back to when fixation had stabilized after the last action. The sequences were then clustered using the longest common sequence method in the TraMineR package of R-project statistical software. The clusters, similar in fixations sequence and length, were then associated with the tactics and actions performed. In total 23 fixation sequences and a same number of actions were identified. The most frequent and longest fixations were concentrated around the body and head of the opponent. The dendrogram showed three major types of clusters. Type 1 was mostly associated with counter attacks (79%) that included punches (57%). The second and third clusters contained the same number of attacks as counters and were mostly associated with lunges (50%) and kicks (38%). Furthermore looking at the lengths of the sequences within each cluster, type 2 and 3 had longer lengths than type 1. Therefore, these two clusters could essentially be considered as one despite differences in eye fixation sequence. The counter chudan (punch to the body) technique was used the most (7 times) and is associated with cluster type 1. Attacking kicks (5 times) and backward lunges (5 times) on the other hand are associated with the combination of type 2 and 3. However, it is rather the length of the fixation sequence in between the attacks than the area of fixations themselves that was found to be associated with the actions by this karateka. Short fixation sequences evoke a fast punching response. This is likely due to pressured situations where the distance to the opponent was reduced quickly by an attacking opponent. While longer fixation sequences are followed by (evading) lunges and kicks to disrupt the opponent and keep distance from the opponent without a direct attempt to score.

Registro de presencia y actividades de algunos mamíferos en el Desierto Chihuahuense, México

Introduction: Photographic records are useful tools that had strengthen wildlife mammal studies as a noninvasive methodology providing a wide variety of information. The objective of this study is to document with images (automatic - trail cameras and manually – personal cameras) the species of mammals inhabiting the Mapimi Biospere Reserve (MBR), to highlight the morphologic characteristics that distinguish them, to record the area and vegetation type where each species was found and to determine some of their daily activities. Methods: Between 2009 and 2014, twelve field trips were made to Mapimi. In each trip we placed six trail cameras each at five locations (30 cameras total) for 24 h during six consecutive days, with 51,840 h photographic effort. Two sets of six cameras were placed in grassland habitat and three in shrubland habitat. Each camera was baited with 300 g of alfalfa mixed with sand in a tray placed 3 m from the camera. Between 2011 and 2014 two additional cameras were set to monitor a natural pond on the slope of a small mountain (San Ignacio) in the area with 5,760 h of photographic effort. In total we had 57,600 h of photographic effort. Images were also obtained from manual cameras of the authors while engaged in various research projects in the area with more than 2,000 photographs. These data for some species are directly from animals captured in the locality. Their main morphologic characteristics were highlighted, specific places, type of vegetation where each species was found and some of their daily activities. Results: A total number of five orders, 12 families, 27 genera and 31 species were recorded, representing 49 % of mammals known in the Reserve. With trail cameras 1,714 photos were taken and 15 species were recorded (Table 1). The species with highest number of records were Dipodomys nelsoni ( n = 610), D. merriami ( n = 469) and Netoma leucodon ( n = 335). The ones with the lowest number of photos were Ictidomys mexicanus ( n = 48) and Sylvilagus audubonii ( n = 46, Figure 1).Trail cameras recorded five genera where the species level could not be determined: Myotis sp., Chaetodipus sp. , Peromyscus sp., Ictidomys or Xerospemorphilus . With manual cameras more than 2,000 photos were taken and thirty species were photographed, only 16 were recorded with this method (Table 1). The collared peccary ( Pecari tajacu) was the single species recorded only by trailed cameras. Discussion and Conclusions: Results obtained demonstrated that with trail cameras, morphological characteristics that distinguish most mammalian species can be observed clearly for most small, medium and large species. For some small species (such as Heteromyidae and Muridae) their features cannot be easily noted so their proper identity is difficult to determine. It is also important to discuss that only half of the known mammal diversity in the locality was recorded with trail cameras. Some small (such as Heteromyidae and Muridae rodents and bats), medium (as skunks) or large size species (as pumas) were not recorded with them, even though these cameras are generally considered a very effective method for mammalian inventories. As our results do not support this idea, we strongly recommend to combine direct trapping methods with trail cameras to get more complete and more realistic mammalian inventories. The highest numbers of records obtained with trail cameras were of Dipodomys nelsoni , D. merriami and Netoma leucodon , reflecting their numerical dominance in the ecosystem. Lower number of records of Sylvilagus audubonii, squirrels, deer , peccaries and other mammals are related with their low abundance and they occur only in very specific areas. The low numbers found of Lepus californicus were contradictory with previous studies, where they were much more frequently recorded. Skunks inhabit rocky and hills parts in the Reserve, and the presence of Spilogale gracilis was exceptional and as such represents a new record of this species for the Reserve. Documentation of mammalian species through photographic images contributes to different aspects of their biology that had not been addressed before in this location. Cameras allowed updating of information of some species and helped to identify knowledge gaps that can be addressed in the future to contribute to understand the dynamics of the mammal community in the Chihuahuan Desert. Key words: Manual cameras, new record, photographic identification, Spilogale gracilis , trail cameras.

Robust action recognition via borrowing information across video modalities.

The recent advances in imaging devices have opened the opportunity of better solving the tasks of video content analysis and understanding. Next-generation cameras, such as the depth or binocular cameras, capture diverse information, and complement the conventional 2D RGB cameras. Thus, investigating the yielded multimodal videos generally facilitates the accomplishment of related applications. However, the limitations of the emerging cameras, such as short effective distances, expensive costs, or long response time, degrade their applicability, and currently make these devices not online accessible in practical use. In this paper, we provide an alternative scenario to address this problem, and illustrate it with the task of recognizing human actions. In particular, we aim at improving the accuracy of action recognition in RGB videos with the aid of one additional RGB-D camera. Since RGB-D cameras, such as Kinect, are typically not applicable in a surveillance system due to its short effective distance, we instead offline collect a database, in which not only the RGB videos but also the depth maps and the skeleton data of actions are available jointly. The proposed approach can adapt the interdatabase variations, and activate the borrowing of visual knowledge across different video modalities. Each action to be recognized in RGB representation is then augmented with the borrowed depth and skeleton features. Our approach is comprehensively evaluated on five benchmark data sets of action recognition. The promising results manifest that the borrowed information leads to remarkable boost in recognition accuracy.