Background Motion Research Articles

Movement Disorders and Hematologic Diseases.

Movement disorders can be associated with or caused by hematological abnormalities. The objective of this review is to highlight features that will aid in the clinician's recognition and treatment of these disorders. MESH terms relevant to movement disorders and hematologic diseases were searched to identify conditions included in this narrative, educational review. Several conditions were identified, and they were organized by hematologic categories to include red blood cell abnormalities, white blood cell abnormalities, disorders of clotting and bleeding, hematologic malignancies, and others. This review will increase providers' understanding of disorders that include movement disorders and hematologic abnormalities. Basic hematologic laboratories can aid in assessment of these disorders, to include complete blood count/hemogram and peripheral blood smear. Recognition is key, especially in the setting of underlying malignancy, vitamin deficiency, or other disorder in which treatment is available.

Visually guided inspiration breath-hold facilitated with nasal high flow therapy in locally advanced lung cancer

Background and purpose Reducing breathing motion in radiotherapy (RT) is an attractive strategy to reduce margins and better spare normal tissues. The objective of this prospective study (NCT03729661) was to investigate the feasibility of irradiation of non-small cell lung cancer (NSCLC) with visually guided moderate deep inspiration breath-hold (IBH) using nasal high-flow therapy (NHFT). Material and methods Locally advanced NSCLC patients undergoing photon RT were given NHFT with heated humidified air (flow: 40 L/min with 80% oxygen) through a nasal cannula. IBH was monitored by optical surface tracking (OST) with visual feedback. At a training session, patients had to hold their breath as long as possible, without and with NHFT. For the daily cone beam CT (CBCT) and RT treatment in IBH, patients were instructed to keep their BH as long as it felt comfortable. OST was used to analyze stability and reproducibility of the BH, and CBCT to analyze daily tumor position. Subjective tolerance was measured with a questionnaire at 3 time points. Results Of 10 included patients, 9 were treated with RT. Seven (78%) completed the treatment with NHFT as planned. At the training session, the mean BH length without NHFT was 39 s (range 15–86 s), and with NHFT 78 s (range 29–223 s) (p = .005). NHFT prolonged the BH duration by a mean factor of 2.1 (range 1.1–3.9s). The mean overall stability and reproducibility were within 1 mm. Subjective tolerance was very good with the majority of patients having no or minor discomfort caused by the devices. The mean inter-fraction tumor position variability was 1.8 mm (-1.1–8.1 mm;SD 2.4 mm). Conclusion NHFT for RT treatment of NSCLC in BH is feasible, well tolerated and significantly increases the breath-hold duration. Visually guided BH with OST is stable and reproducible. We therefore consider this an attractive patient-friendly approach to treat lung cancer patients with RT in BH.

Two-phase Motion in Hydrodynamic Counter-current Chromatography.

Background Motion of the two mutually immiscible liquids in hydrodynamic countercurrent chromatographic systems is speculated based on the observation of their behavior in a closed coiled tube rotating in unit gravity.Materials and Methods The experiment revealed an up and down pattern of four stages of two-phase volume ratio occupied at the head end of the coil according to the rotation speed. These two-phase behaviors are comprehensively explained on the bases of interplay between the unit gravity and centrifugal force generated by rotation of the coil. This theory is successfully extended to explain the two-phase behavior in a coil undergoing the type-I and type-J planetary motions.Results and Discussion The type-I planetary motion produces the centrifugal force distribution similar to that of slowly rotating coil in unit gravity (Stage I), where both phases competitively move toward the head of the coil. In contrast, the type-J planetary motion displays complex distribution patterns of centrifugal force according to the location of the coil on the holder hence the two-phase motion varies with the ß values. When ß is 0.5 – 0.75, the force pattern simulates that of the rotating coil in unit gravity at 120 rpm (Stage III) where the lighter phase moves toward the head leaving the heavier phase behind.Conclusion This clearly demonstrates the importance of the proper choice of ß values in high-speed countercurrent chromatography utilizing the type-J planetary motion.

Surface guided motion management in glottic larynx stereotactic body radiation therapy

Background and purposeInvoluntary motion due to swallowing cause inaccurate dose delivery during larynx radiotherapy, a deviation that may be particularly problematic during stereotactic body radiation therapy (SBRT). The goal of this study was to develop a motion management solution for larynx SBRT using surface imaging. Material and methodsTen patients were recently treated on a phase II study of larynx SBRT on a LINAC equipped with a surface guidance system. A small region of the immobilization mask was manually cut open to allow surface tracking. Pre-treatment and intra-fractional CBCTs were acquired to verify internal anatomy. Patients were verbally instructed not to swallow during treatment. During treatment delivery, beam hold was initiated by the Motion Management Interface if surface motion exceeded a patient-specific threshold. Patient motion was recorded in log files and analyzed. We also performed phantom studies to assess the theoretical impact of gating on dose delivery. ResultsThe frequency (6.5 ± 5.2 times per fraction) and duration (3.9 ± 2.5 seconds per swallow) of swallowing varied both between patients and fractions. The magnitude of each swallow showed mean peak amplitude at 5.8 ± 3.8 mm above baseline, mostly in the longitudinal direction. Beam duty cycle was 95.0% ± 7.0% (absolute range: 76–100%), with inefficiency most prominent in the early fractions. The 95th percentile residual motion was reduced from 3.4 mm to 2.3 mm with both verbal instruction and gating. Phantom studies confirmed dose delivery accuracy represented by gamma pass rate was improved by 5% using this approach. ConclusionsLaryngeal motion management using surface imaging is feasible and efficacious. Uncontrolled movement of the larynx was not uncommon during treatment, with gating reducing potential for unplanned dose deviations. Additional research is needed to determine the clinical benefit with this system.

Upsampling Real-Time, Low-Resolution CCTV Videos Using Generative Adversarial Networks

Video super-resolution has become an emerging topic in the field of machine learning. The generative adversarial network is a framework that is widely used to develop solutions for low-resolution videos. Video surveillance using closed-circuit television (CCTV) is significant in every field, all over the world. A common problem with CCTV videos is sudden video loss or poor quality. In this paper, we propose a generative adversarial network that implements spatio-temporal generators and discriminators to enhance real-time low-resolution CCTV videos to high-resolution. The proposed model considers both foreground and background motion of a CCTV video and effectively models the spatial and temporal consistency from low-resolution video frames to generate high-resolution videos. Quantitative and qualitative experiments on benchmark datasets, including Kinetics-700, UCF101, HMDB51 and IITH_Helmet2, showed that our model outperforms the existing GAN models for video super-resolution.

Self-supervised video representation learning by maximizing mutual information

We address the problem of learning representations from the videos without manual annotation. Different video clips sampled from the same video usually have a similar background and consistent motion. A novel self-supervised task is designed to learn such temporal coherence, which is measured by the mutual information in our work. First, we maximize the mutual information between features extracted from the clips which are sampled from the same video. This encourages the network to learn the shared content by these clips. As a result, the network may focus on the background and ignore the motion in videos due to that different clips from the same video normally have the same background. Second, to address this issue, we simultaneously maximize the mutual information between the feature of the video clip and the local regions where salient motion exists. Our approach, which is referred to as Deep Video Infomax (DVIM), strikes a balance between the background and the motion when learning the temporal coherence. We conduct extensive experiments to test the performance of the proposed DVIM on various tasks. Experimental results of fine-tuning for the high-level action recognition problems validate the effectiveness of the learned representations. Additional experiments for the task of action similarity labeling also demonstrate the generalization of the learned representations of the DVIM.

Movement, posture and low back pain. How do they relate? A replicated single-case design in 12 people with persistent, disabling low back pain.

Movement and posture are commonly believed to relate to non-specific low back pain (NSLBP). While people with NSLBP appear to move and posture themselves differently from those without NSLBP, changes in movement and posture infrequently relate to improvements in NSLBP when analysed at a group-level. Additionally, little is known about how movement or posture change when clinical outcome improves. Within-person relationships were investigated using a replicated, repeated measures, single-case design in 12 people with persistent, disabling NSLBP. Individually relevant movement and posture were captured using wearable sensors on up to 20 occasions over a 22-week period (5-week baseline, 12-week physiotherapy-led intervention, 5-week follow-up), while pain and activity limitation were collected concomitantly. A series of cross-correlation analyses estimated the presence, strength, and direction of relationships. Many participants (n=10/12) had strong (e.g. r=0.91, p=<0.001) relationships between changes in movement or posture and changes in pain and activity limitation, while some showed no strong association. Where relationships were observed, clinical improvement predominantly (93% or 57/61 relationships) related to increased spinal movement range and velocity during forward bending and lifting, reduced lumbar muscle EMG activity at maximum voluntary flexion, and increased posterior-pelvic-tilt during sitting and standing. Within-person changes to individually relevant movement and posture appear to often relate to clinical outcome, but not always. When changes were related, movement and posture appear to return towards being 'less protective', however causal directions remain unknown. Important activities, movements, and postural parameters varied across the participants, highlighting the potential importance of individualized management. Changes to individually relevant movement and posture appear to often relate to clinical outcome, but not always. Patient-specific activities, and movement or postural parameters that related to improved pain and activity limitation, varied across the 12 participants, highlighting the potential importance of individualised management. Where clinical improvements were related to changes in movement or posture, participants consistently returned towards being 'less protective' (increased range and speed of movement, increased posterior-pelvic-tilt during sitting and standing). Mechanisms and generalizability remain unclear.

Modelling Drosophila motion vision pathways for decoding the direction of translating objects against cluttered moving backgrounds

Decoding the direction of translating objects in front of cluttered moving backgrounds, accurately and efficiently, is still a challenging problem. In nature, lightweight and low-powered flying insects apply motion vision to detect a moving target in highly variable environments during flight, which are excellent paradigms to learn motion perception strategies. This paper investigates the fruit fly Drosophila motion vision pathways and presents computational modelling based on cutting-edge physiological researches. The proposed visual system model features bio-plausible ON and OFF pathways, wide-field horizontal-sensitive (HS) and vertical-sensitive (VS) systems. The main contributions of this research are on two aspects: (1) the proposed model articulates the forming of both direction-selective and direction-opponent responses, revealed as principal features of motion perception neural circuits, in a feed-forward manner; (2) it also shows robust direction selectivity to translating objects in front of cluttered moving backgrounds, via the modelling of spatiotemporal dynamics including combination of motion pre-filtering mechanisms and ensembles of local correlators inside both the ON and OFF pathways, which works effectively to suppress irrelevant background motion or distractors, and to improve the dynamic response. Accordingly, the direction of translating objects is decoded as global responses of both the HS and VS systems with positive or negative output indicating preferred-direction or null-direction translation. The experiments have verified the effectiveness of the proposed neural system model, and demonstrated its responsive preference to faster-moving, higher-contrast and larger-size targets embedded in cluttered moving backgrounds.

Accurate moving object segmentation in unconstraint videos based on robust seed pixels selection

Due to the clutter background motion, accurate moving object segmentation in unconstrained videos remains a significant open problem, especially for the slow-moving object. This article proposes an accurate moving object segmentation method based on robust seed selection. The seed pixels of the object and background are selected robustly by using the optical flow cues. Firstly, this article detects the moving object’s rough contour according to the local difference in the weighted orientation cues of the optical flow. Then, the detected rough contour is used to guide the object and the background seed pixel selection. The object seed pixels in the previous frame are propagated to the current frame according to the optical flow to improve the robustness of the seed selection. Finally, we adopt the random walker algorithm to segment the moving object accurately according to the selected seed pixels. Experiments on publicly available data sets indicate that the proposed method shows excellent performance in segmenting moving objects accurately in unconstraint videos.

Comparison of different methods for lung immobilization in an animal model

Background and purposeRespiratory-induced motion introduces uncertainties in the delivery of dose in radiotherapy treatments. Various methods are used clinically, e.g. breath-holding, while there is limited experience with other methods such as apneic oxygenation and high frequency jet ventilation (HFJV). This study aims to compare the latter approaches for lung immobilization and their clinical impact on gas exchange in an animal model. Materials and methodsTwo radiopaque tumor surrogate markers (TSM) were placed in the central (cTSM) and peripheral (dTSM) regions of the lungs in 9 anesthetized and muscle relaxed pigs undergoing 3 ventilatory interventions (1) HFJV at rates of 200 (JV200), 300 (JV300) and 400 (JV400) min−1; (2) apnea at continuous positive airway pressure (CPAP) levels of 0, 8 and 16 cmH2O; (3) conventional mechanical ventilation (CMV) as reference mode. cTSM and dTSM were visualized using fluoroscopy and their coordinates were computed. The ventilatory pattern was registered, and oxygen and carbon dioxide (pCO2) partial pressures were measured. ResultsThe highest range of TSM motion, and ventilation was found during CMV, the lowest during apnea. During HFJV the amount of motion varied inversely with increasing frequency. The reduction of TSM motion at JV300, JV400 and all CPAP levels came at the cost of increased pCO2, however the relatively low frequency of 200 min−1 for HFJV was the only ventilatory setting that enabled adequate CO2 removal. ConclusionIn this model, HFJV at 200 min−1 was the best compromise between immobilization and gas exchange for sessions of 10-min duration.

Foreground Extraction Using Hilbert-Schmidt Independence Criterion and Particle Swarm Optimization Independent Component Analysis

Foreground extraction is one of the crucial subjects in image processing, which drives different applications in industry. The reality behind the continuous research in this area is the various challenging problems we encounter during the separation process of foreground and background images. Among the source separation approaches, the independent component analysis (ICA) is the most prevalent, being involved in different areas of signal separation applications. Despite the improvements being achieved in foreground extraction, the sudden luminance variations and background movements adversely impact the results of techniques in this regard. In this paper, a novel structure called HSIC_ICA is introduced to address the mentioned problem using a modified version of the ICA algorithm which, leverages the Hilbert-Schmidt Independence Criterion (HSIC) instead of the common objective functions. Moreover, the unmixing matrix elements of ICA are extracted through a Particle Swarm Optimization (PSO) evolutionary algorithm in a much faster way. The experimental results clearly show that the proposed method outperforms over the significant works being cited among the references, using Wallflower dataset.

Non-canonical Receptive Field Properties and Neuromodulation of Feature-Detecting Neurons in Flies

Several fundamental aspects of motion vision circuitry are prevalent across flies and mice. Both taxa segregate ON and OFF signals. For any given spatial pattern, motion detectors in both taxa are tuned to speed, selective for one of four cardinal directions, and modulated by catecholamine neurotransmitters. These similarities represent conserved, canonical properties of the functional circuits and computational algorithms for motion vision. Less is known about feature detectors, including how receptive field properties differ from the motion pathway or whether they are under neuromodulatory control to impart functional plasticity for the detection of salient objects from a moving background. Here, we investigated 19 types of putative feature selective lobula columnar (LC) neurons in the optic lobe of the fruit fly Drosophila melanogaster to characterize divergent properties of feature selection. We identified LC12 and LC15 as feature detectors. LC15 encodes moving bars, whereas LC12 is selective for the motion of discrete objects, mostly independent of size. Neither is selective for contrast polarity, speed, or direction, highlighting key differences in the underlying algorithms for feature detection and motion vision. We show that the onset of background motion suppresses object responses by LC12 and LC15. Surprisingly, the application of octopamine, which is released during flight, reverses the suppressive influence of background motion, rendering both LCs able to track moving objects superimposed against background motion. Our results provide a comparative framework for the function and modulation of feature detectors and new insights into the underlying neuronal mechanisms involved in visual feature detection.

Fly Feature Detectors Show Contrast Invariance, Omni-Directionality, Velocity Constancy, and Octopaminergic Loss of Background Motion Suppression

Motion vision circuits of the fly are organized into parallel ON and OFF pathways tuned to the direction and speed of motion. Directionally selective small-field columnar neurons innervate the third optic ganglion, the lobula plate, where their outputs are integrated by large dendritic fields that encode wide-field background motion. Wide-field lobula plate cells, in turn, coordinate gaze stabilization. By contrast, behavioral responses to visual features is believed to begin within the circuits of the fourth optic ganglion, the lobula. We investigated 19 lobula columnar neurons (LCs) with stimuli that are behaviorally salient to the fly. A comprehensive analysis of receptive field properties showed that LC15 encodes moving bars, whereas LC12 is relatively agnostic to object dimensions. Neither responds to optic flow, rather object responses in both LCs are suppressed by background motion. Further contrasting neurons of the motion pathway, LC12 and LC15 are both small-field cells indifferent to ON-OFF contrast polarity. LC15 responds equally to four directions of motion, whereas LC12 is selective for motion in the horizontal plane. Both are equally sensitive to object speeds ranging from 22 to 132° per second. Small objects strongly stimulate presynaptic fibers of lobula layer 2, which corresponds to the postsynaptic domains of object-detecting LCs. Finally, in direct contrast to the entire motion pathway, we were surprised to find that octopamine nullifies the inhibition of object responses generated by background motion; LC12 and LC15 remain sensitive to objects appearing against a moving background.

RETRACTED ARTICLE: A kernel support vector machine based anomaly detection using spatio-temporal motion pattern models in extremely crowded scenes

Millions of security cameras were placed in public spaces, generating large quantities of video data. There is a need to develop smart techniques to identify and classify objects tracking instantly. Most of them concentrate on spatial information, resulting in exposure to noise and background movement. In addition, monitoring individuals in overcrowded scenes is a difficult task, due to the variation of movement and appearance created by the large amount of people in the scene. In this paper, initially, utilizing threshold value, the video is split into frames. Then segmentation of moving objects using Extended Kalman Filters (EKF) to improve the accuracy of the classification. Instead, to distinguish between the foreground object and the background object, the texture features is removed. The artifacts are then labelled using improved Learning Vector Quantization (LVQ) for efficient identification of anomalies. Also an effective classification of Kernel Support Vector Machine (KSVM) predicated on anomaly detection has been suggested utilizing spatio-temporal movement pattern models in overcrowded scenes to solve these problems. Hence, KSVM is more advantageous in accuracy which is used to monitor the object. The result shows the performance of the proposed KSVM obtained high performance compared with SVM and Hidden Markov Model (HMM).

Ultrasound diaphragmatic manual and semi-automated motion measurements: Application in simulated and in vivo data of critically ill subjects

Background and ObjectiveUltrasound diaphragmatic muscle motion characteristics may provide useful information about normal or abnormal diaphragmatic function and indicate diaphragmatic weakness, or paralysis. In the present work we propose and evaluate an integrated semi-automated analysis system for the quantitative analysis of ultrasonic motion from ultrasound diaphragmatic videos. MethodsThe proposed system was evaluated in simulated videos and in 13 patients, four of whom patients were mechanically ventilated. The major steps of the methodology were as follows: video normalization, despeckle filtering, generation of an M-Mode image, snakes segmentation, and motion measurements. ResultsThe following manual (-/) vs semi-automated (/-), (median±IQR) measurements, which are routinely carried out by the experts, for assessing the severity of the disease, were computed. For the simulated videos the diaphragmatic excursion was 1.80±0.00 cm / 1.76±0.03 cm. For all the real ultrasound videos investigated in this study the following measurements were computed: (i) diaphragmatic excursion: 0.84±0.15 cm / 0.83±0.14 cm, (ii) inspiration time (Tinsp): 0.71±0.18 sec / 0.70±0.15 sec, (iii) total breathing time for one cycle (Ttot): 1.71±0.37 sec / 1.67±0.37 sec, (iv) diaphragmatic curve slope: 1.29±0.36 cm/sec / 1.27±0.36 cm/sec, and (v) relaxation rate (RR): 0.82±0.17 cm/sec / 0.82±0.18 cm/sec. ConclusionsManual and semi-automated measurements were very close with non-statistical significant differences and strong correlations between them. It is anticipated that the proposed system might be useful in the clinical practice in the assessment and follow up of patients with diaphragmatic weakness or paralysis and aid in the separation of normal and abnormal diaphragmatic motion. Further validation and additional experimentation in a larger sample of videos and different patient groups is required.

Step by step and frame by frame – Workflow for efficient motion tracking of high-speed movements in animals

Analysing the motion of animals, especially at high speeds, is often challenging. Motion tracking software needs to deal with a variety of visual contexts, variable lighting conditions, heterogeneous backgrounds and even background movements. Here we present motion tracking via the easy to use and constantly updated Adobe After Effects software - which is often included in software packages most researchers are already using. The provided custom-made Javascript allows for easy exporting of tracking coordinates. Furthermore, some examples for analysing the obtained data in the open source statistical software 'R' will provide reference points, even for an unexperienced user. We present a step-by-step guide of the methodology using high-speed video recordings of locust jumps and additionally validate this method by successful tracking of simulated data under defined subpar filming conditions. This simulated data allows experienced users to compare the tracking software in use with the here presented workflow to weigh the advantages and disadvantages of any motion tracking software on the market.

Side of contrast injection and breast size correlate with motion artifacts grade and image quality on breast MRI.

Motion is a relevant cause of artifacts in breast magnetic resonance imaging (MRI), potentially degrading image quality, even with optimized protocols. To investigate the causes of motion artifacts (MA) impacting on image quality (IQ) of contrast-enhanced breast MRI. Retrospective two-center study on consecutive 1.5-T contrast-enhanced breast MRI, independently reviewed by two radiologists on first subtracted and maximum intensity projection images to define the side most affected by MA. IQ was scored as 1 (optimal), 2 (reduced, but without reduction of diagnostic power), or 3 (reduced, with reduced diagnostic power). Correlations with injection side, breast size (A/B vs. C/D cups), patient age, clinical indication, and MRI scanner/protocol were assessed using χ2 and Fisher's exact statistics. In total, 237 examinations were included, with right injection performed in 124 (52%) and left in 113 (48%). MA were more frequent on the side ipsilateral to the injection (144/237, 61%, 95% confidence interval [CI] 54-67%) than on the contralateral (93/237, 95% CI 33-46%) (P < 0.001); IQ was scored 1 in 154/237 (65%), 2 in 63/237 (27%), and 3 in 20/237 (8%) examinations; patients with A/B cups showed higher IQ score than patients with C/D cups (scores 1, 2, and 3: 54% vs. 70%; 29% vs. 25%; 17% vs. 5%, respectively, P = 0.002). No significant correlations were found for MA (P≥0.106) or IQ (P ≥ 0.318) between ipsi- or contralateral injection and right/left injection, breast size, age, indication, or scanner/protocol. MA were more frequent in breasts ipsilateral to contrast injection and showed a reduced IQ for small breasts.

Prospects for the Use of Upconverting Nanoparticles as a Contrast Agent for Enumeration of Circulating Cells in vivo.

PurposeWe recently developed a new fluorescence-based technique called “diffuse in vivo flow cytometry” (DiFC) for enumerating rare circulating tumor cells (CTCs) directly in the bloodstream. Non-specific tissue autofluorescence is a persistent problem, as it creates a background which may obscure signals from weakly-labeled CTCs. Here we investigated the use of upconverting nanoparticles (UCNPs) as a contrast agent for DiFC, which in principle could significantly reduce the autofluorescence background and allow more sensitive detection of rare CTCs.MethodsWe built a new UCNP-compatible DiFC instrument (U-DiFC), which uses a 980 nm laser and detects upconverted luminescence in the 520, 545 and 660 nm emission bands. We used NaYF4:Yb,Er UCNPs and several covalent and non-covalent surface modification strategies to improve their biocompatibility and cell uptake. We tested U-DiFC with multiple myeloma (MM) and Lewis lung carcinoma (LLC) cells in tissue-mimicking optical flow phantoms and in nude mice.ResultsU-DiFC significantly reduced the background autofluorescence signals and motion artifacts from breathing in mice. Upconverted luminescence from NaYF4:Yb,Er microparticles (UμNP) and cells co-incubated with UCNPs were readily detectable with U-DiFC in phantoms, and from UCNPs in circulation in mice. However, we were unable to achieve reliable labeling of CTCs with UCNPs. Our data suggest that most (or all) of the measured U-DIFC signal in vitro and in vivo likely arose from unbound UCNPs or due to the uptake by non-CTC blood cells.ConclusionUCNPs have a number of properties that make them attractive contrast agents for high-sensitivity detection of CTCs in the bloodstream with U-DiFC and other intravital imaging methods. More work is needed to achieve reliable and specific labeling of CTCs with UCNPs and verify long-term retention and viability of cells.

Learning 3D spatiotemporal gait feature by convolutional network for person identification

For person identification in non-interaction biometric systems, gait recognition has been recently encouraged in literature and industrial applications instead of face recognition. Although numerous advanced methods that learn object appearance by conventional machine learning models have been discussed in the last decade, most of them are strongly sensitive to scene background motion. In this research, we address the drawbacks of existing works by comprehensively studying gait information from 3D human skeleton data with a deep learning-based identifier. To capture the statistic gait information in the spatial dimension, we first extract the geometric gait features of joint distance and orientation. The dynamic gait information is then obtained by calculating the temporal description features with the mean and standard deviation of geometric features. Accordingly, the fully gait information of an individual is finally learned via a compact Deep Convolutional Neural Network which is explicitly designed with multiple stacks of asymmetric convolutional filters to fully gain the spatial correlation of in-frame body joints and the temporal relation of frame-wise posture at multiple scales. Based on the experimental results evaluated on four benchmark 3D gait datasets commonly used for person identification, including UPCV Gait, UPCV Gait K2, KS20 VisLab Multi-View Kinect Skeleton, and SDUGait, the proposed method presents the superior performance over that of several state-of-the-art approaches while maintaining a low computational capacity.

Inhibitory Interactions and Columnar Inputs to an Object Motion Detector in Drosophila

The direction-selective T4/T5 cells innervate optic-flow processing projection neurons in the lobula plate of the fly that mediate the visual control of locomotion. In the lobula, visual projection neurons coordinate complex behavioral responses to visual features, however, the input circuitry and computations that bestow their feature-detecting properties are less clear. Here, we study a highly specialized small object motion detector, LC11, and demonstrate that its responses are suppressed by local background motion. We show that LC11 expresses GABA-A receptors that serve to sculpt responses to small objects but are not responsible for the rejection of background motion. Instead, LC11 is innervated by columnar T2 and T3 neurons that are themselves highly sensitive to small static or moving objects, insensitive to wide-field motion and, unlike T4/T5, respond to both ON and OFF luminance steps.