Search

Recent years have witnessed great progress in creating vivid audio-driven portraits from monocular videos. However, how to seamlessly adapt the created video avatars to other scenarios with different backgrounds and lighting conditions remains unsolved. On the other hand, existing relighting studies mostly rely on dynamically lighted or multi-view data, which are too expensive for creating video portraits. To bridge this gap, we propose ReliTalk, a novel framework for relightable audio-driven talking portrait generation from monocular videos. Our key insight is to decompose the portrait's reflectance from implicitly learned audio-driven facial normals and images. Specifically, we involve 3D facial priors derived from audio features to predict delicate normal maps through implicit functions. These initially predicted normals then take a crucial part in reflectance decomposition by dynamically estimating the lighting condition of the given video. Moreover, the stereoscopic face representation is refined using the identity-consistent loss under simulated multiple lighting conditions, addressing the ill-posed problem caused by limited views available from a single monocular video. Extensive experiments validate the superiority of our proposed framework on both real and synthetic datasets. Our code is released in (https://github.com/arthur-qiu/ReliTalk). [ABSTRACT FROM AUTHOR]

Plant height is an important parameter of plant phenotype as one indicator of plant growth. In view of the complexity and scale limitation in current measurement systems, a scaleless method is proposed for the automatic measurement of plant height based on monocular computer vision. In this study, four peppers planted side by side were used as the measurement objects. Two color images of the measurement object were obtained by using a monocular camera at different shooting heights. Binary images were obtained as the images were processed by super-green grayscale and the Otsu method. The binarized images were transformed into horizontal one-dimensional data by the statistical number of vertical pixels, and the boundary points of multiple plants in the image were found and segmented into single-plant binarized images by filtering and searching for valleys. The pixel height was extracted from the segmented single plant image and the pixel displacement of the height was calculated, which was substituted into the calculation together with the reference height displacement to obtain the realistic height of the plant and complete the height measurements of multiple plants. Within the range of 2–3 m, under the light condition of 279 lx and 324 lx, this method can realize the rapid detection of multi-plant phenotypic parameters with a high precision and obtain more accurate plant height measurement results. The absolute error of plant height measurement is not more than ±10 mm, and the absolute proportion error is not more than ±4%. [ABSTRACT FROM AUTHOR]

Underwater target three-dimensional detection is crucial for effectively recognizing and acquiring target information in complex water rings. The underwater robotic operating system as a conventional underwater operating platform, generally equipped with a binocular or monocular camera, how to utilize the underwater monocular camera with high precision and high efficiency to complete the target three-dimensional information acquisition is the main research starting point of this paper. To this end, this paper proposes a laser-assisted three-dimensional depth monocular detection method for underwater targets, which utilizes three cross lasers to assist the monocular camera system in capturing the depth data at different positions of the target plane at one time. The image correction by the four-point laser calibration method in this paper solves the difficulties of image distortion caused by an unstable underwater environment and lens effect, as well as laser angle deviation caused by the tilting of the underwater robot. The instability of the underwater environment and the lens can cause image distortion, and the tilt of the underwater robot causes the laser angle to deviate. After correcting the image, the depth data between the target and the robot can be calculated based on the geometric relationship that exists between the imaginary rectangle formed by the laser dots and laser lines in the image and the imaginary rectangle formed between the lasers on the device. This method uses a single image to obtain target depth information and is capable of measuring not only horizontal planes but also multiplanes and inclined planes. Experiments show that the algorithm improves the performance accuracy in underwater environments and land environments compared to traditional methods, and obtains depth information for the entire plane at once. The method provides a theoretical and practical basis for underwater monocular 3D information acquisition. [ABSTRACT FROM AUTHOR]

The purpose of this study was to examine how demographics, etiology, and clinical examination findings are related to visual outcomes in subjects with open globe injury (OGI) across a large and generalizable sample. A retrospective cohort analysis was performed using data collected from the electronic medical records of four tertiary university centers for subjects with OGI presenting from 2018 to 2021. Demographic information, injury mechanisms, clinical exam findings, visual acuity (VA) at presentation and most recent follow-up were recorded. In subjects with bilateral OGIs, only right eyes were included. A modified ocular trauma score (OTS) using presenting VA, the presence of perforating injury, OGI, and afferent pupillary defect was calculated. The risk of subjects' demographic characteristics, ocular trauma etiology, clinical findings and modified OTS on the presence of monocular blindness at follow-up were assessed using univariable and multivariable regression models. 1426 eyes were identified. The mean age was 48.3 years (SD: ± 22.4 years) and the majority of subjects were men (N = 1069, 75.0%). Univariable analysis demonstrated that subjects of Black race were 66% (OR: 1.66 [1.25–2.20]; P < 0.001) more likely to have monocular blindness relative to White race at follow-up. OTS Class 1 was the strongest predictor of blindness (OR: 38.35 [21.33–68.93]; P < 0.001). Based on multivariable analysis, lower OTS category (OTS Class 1 OR: 23.88 [16.44–45.85]; P < 0.001) moderately predicted visual outcomes (R2 = 0.275, P < 0.001). OGI has many risks of poor visual outcome across patient groups that vary by demographic category, mechanism of injury, and clinical presentation. Our findings validate that a modified OTS remains a strong predictor of visual prognosis following OGI in a large and generalizable sample. [ABSTRACT FROM AUTHOR]

Road detection is an essential component of indoor robot navigation. Vision sensors have great advantages in road detection as they can provide rich information in terms of environmental perception. In this paper, a monocular vision sensor-based method for indoor road and stair detection is proposed, which detects feasible areas in indoor environments very fast without paying attention to detailed features of walls or other obstacles. More specifically, for a given indoor road image captured by an on-board vision sensor, the simple linear iterative clustering (SLIC) algorithm-based approach for efficient image segmentation is introduced. Then, according to the DBSCAN algorithm, the generated superpixels are clustered to form large areas of view. The initial road area is obtained through a safe window on the middle bottom of the image. In order to achieve a more accurate road segmentation, the initial image is processed by the binary search, edge detection based on the Canny operator and straight-line detection and location based on the Hough transform, which integrates edge and stair information into road detection. Several experiments are performed to evaluate the performance of the proposed method. The experimental results demonstrate that the proposed method could accurately detect road information and staircase information in images and succeeds in addressing the indoor road-detection problem. [ABSTRACT FROM AUTHOR]

Previous literature suggests that correcting ongoing movements is more effective when using the dominant limb and seeing with the dominant eye. Specifically, individuals are more effective at adjusting their movement to account for an imperceptibly perturbed or changed target location (i.e., online movement correction), when vision is available to the dominant eye. However, less is known if visual-motor functions based on monocular information can undergo short-term neuroplastic changes after a bout of practice, to improve online correction processes. Participants (n = 12) performed pointing movements monocularly and their ability to correct their movement towards an imperceptibly displaced target was assessed. On the first day, the eye associated with smaller correction amplitudes was exclusively trained during acquisition. While correction amplitude was assessed again with both eyes monocularly, only the eye with smaller correction amplitudes in the pre-test showed significant improvement in delayed retention. These results indicate that monocular visuomotor pathways can undergo short-term neuroplastic changes. [ABSTRACT FROM AUTHOR]

Retinoblastoma is a rare childhood cancer caused by a mutation in the RB1 gene. Despite decades of research, there is still much to learn about the genetic makeup of the disease. The availability of treatment options varies by country, with underdeveloped countries having limited access to advanced therapies. Genetic testing plays a crucial role in understanding the inheritance and predisposition to retinoblastoma. Newer techniques, such as cell-free DNA analysis, offer promising alternatives for diagnosis and monitoring. Further research is needed to identify biomarkers and improve diagnostic and treatment options. [Extracted from the article]

This review describes various aspects of traumatic cataracts in order to provide a thorough understanding of this condition. Ocular trauma is an important cause of monocular blindness worldwide. Injury to the lens after blunt or penetrating trauma is common and can result in vision impairment. Selecting the most appropriate therapeutic approaches depends on factors such as patients' age, mechanism of trauma, and underlying clinical conditions. Early management, especially within childhood, is essential because of the difficulties involved in examination; anatomical variations; as well as accompanying intraocular inflammation, amblyopia, or vitreoretinal adhesions. The objective of this study was to provide a comprehensive review of the epidemiology and clinical management of traumatic cataract, highlighting the significance of accurate diagnosis and selection of the optimal therapeutic approach. [ABSTRACT FROM AUTHOR]

Visual ranging technology holds great promise in various fields such as unmanned driving and robot navigation. However, complex dynamic environments pose significant challenges to its accuracy and robustness. Existing monocular visual ranging methods are susceptible to scale uncertainty, while binocular visual ranging is sensitive to changes in lighting and texture. To overcome the limitations of single visual ranging, this paper proposes a fusion method for monocular and binocular visual ranging based on an adaptive Unscented Kalman Filter (AUKF). The proposed method first utilizes a monocular camera to estimate the initial distance based on the pixel size, and then employs the triangulation principle with a binocular camera to obtain accurate depth. Building upon this foundation, a probabilistic fusion framework is constructed to dynamically fuse monocular and binocular ranging using the AUKF. The AUKF employs nonlinear recursive filtering to estimate the optimal distance and its uncertainty, and introduces an adaptive noise-adjustment mechanism to dynamically update the observation noise based on fusion residuals, thus suppressing outlier interference. Additionally, an adaptive fusion strategy based on depth hypothesis propagation is designed to autonomously adjust the noise prior of the AUKF by combining current environmental features and historical measurement information, further enhancing the algorithm's adaptability to complex scenes. To validate the effectiveness of the proposed method, comprehensive evaluations were conducted on large-scale public datasets such as KITTI and complex scene data collected in real-world scenarios. The quantitative results demonstrate that the fusion method significantly improves the overall accuracy and stability of visual ranging, reducing the average relative error within an 8 m range by 43.1% and 40.9% compared to monocular and binocular ranging, respectively. Compared to traditional methods, the proposed method significantly enhances ranging accuracy and exhibits stronger robustness against factors such as lighting changes and dynamic targets. The sensitivity analysis further confirmed the effectiveness of the AUKF framework and adaptive noise strategy. In summary, the proposed fusion method effectively combines the advantages of monocular and binocular vision, significantly expanding the application range of visual ranging technology in intelligent driving, robotics, and other fields while ensuring accuracy, robustness, and real-time performance. [ABSTRACT FROM AUTHOR]

This article discusses a case report that highlights the importance of visual field testing in diagnosing neurological conditions. The case involves a patient with antiphospholipid syndrome and a cerebral vascular malformation who presented with neurological visual field loss. The article emphasizes the need for communication between different medical disciplines in managing patients with conflicting systemic diagnoses. The patient's visual field testing revealed a left superior paracentral scotoma in both eyes, indicating new-onset intracranial pathology in the right occipital lobe. An MRI confirmed the presence of a stable cavernous malformation and a new occipital lobe infarct. The patient was admitted to the hospital for further care. The document also provides education guidelines and learning objectives for optometry students, emphasizing the importance of visual field testing in identifying neurological field loss and differentiating conditions for appropriate referral. Optometrists can play a crucial role in diagnosing and managing these conditions and communicating with other healthcare providers. [Extracted from the article]

Optic perineuritis is an inflammatory condition that presents with reduced visual acuity and painful eye movement. The presentation of optic perineuritis is similar of optic neuritis which results in delayed diagnosis and management. Up to this date, we found a single case of optic neuritis that presented with transient monocular vision loss (TMVL). No cases of optic perineuritis were associated with TMVL. Here, we report a case of a 30-year-old woman who presented with recurrent attacks of painless vision loss in her left eye, reaching up to 30 attacks per day. Ophthalmological examination was otherwise unremarkable. Lab investigations were normal. Magnetic resonance imaging was done, which showed left optic nerve sheath enhancement suggestive of left-sided focal optic perineuritis. Patient was managed with 1 mg IV methylprednisolone for 3 days. We report this case to shed light on the importance of accurate and early diagnosis of optic perineuritis presenting with TMVL. Prompt management of optic perineuritis is crucial in reducing morbidity and risk of relapse. [ABSTRACT FROM AUTHOR]

Purpose: To assess visual and refractive outcomes and visual function after bilateral RayOne Trifocal toric and nontoric intraocular lens (IOL) (Rayner) implantation in patients with presbyopia. Methods: Charts of patients with presbyopia who underwent refractive lens exchange with bilateral implantation of the RayOne Trifocal IOL (toric and non-toric) were retrospectively reviewed. Visual and refractive outcomes were evaluated at 3 months. Patient satisfaction, spectacle independence, and visual disturbance profile were assessed by questionnaires. Results: A total of 5,226 patients were assigned to one of two groups: 1,010 patients had toric IOL implantation (toric group) and 4,216 patients received the non-toric model (non-toric group). Mean ± standard deviation visual acuity at 3 months for the toric group was binocular uncorrected distance visual acuity (UDVA) of 0.07 ± 0.11 logMAR, monocular corrected distance visual acuity (CDVA) of 0.05 ± 0.07 logMAR, binocular uncorrected near visual acuity (UNVA) at 40 cm of 0.10 ± 0.09 logMAR, binocular uncorrected intermediate visual acuity (UIVA) at 40 cm of 0.13 ± 0.12 logMAR, postoperative spherical equivalent (SE) of −0.21 ± 0.47 diopters (D), and cylinder of −0.34 ± 0.40 D. The non-toric group had binocular UDVA of 0.04 ± 0.08 logMAR, monocular CDVA of 0.05 ± 0.07 logMAR, binocular UNVA of 0.10 ± 0.08 logMAR, binocular UIVA of 0.13 ± 0.11 logMAR, SE of −0.08 ± 0.38 D, and cylinder of −0.28 ± 0.34 D. No statistically significant differences were found in achieving spectacle independence and there were high levels of satisfaction in both groups. Conclusions: In this retrospective analysis with more than 5,000 patients, both the toric and non-toric RayOne Trifocal IOL models provided good visual performance at all distances, resulting in excellent levels of spectacle independence and patient satisfaction. [J Refract Surg. 2024;40(7):e468–e479.] [ABSTRACT FROM AUTHOR]

The positioning error of the GPS method at close distances is relatively large, rendering it incapable of accurately guiding unmanned surface vehicles (USVs) back to the mother ship. Therefore, this study proposes a near-distance recovery method for USVs based on monocular vision. By deploying a monocular camera on the USV to identify artificial markers on the mother ship and subsequently leveraging the geometric relationships among these markers, precise distance and angle information can be extracted. This enables effective guidance for the USVs to return to the mother ship. The experimental results validate the effectiveness of this approach, with positioning distance errors of less than 40 mm within a 10 m range and positioning angle errors of less than 5 degrees within a range of ±60 degrees. [ABSTRACT FROM AUTHOR]

Background: Open globe injuries (OGIs) are a leading cause of monocular blindness worldwide and require prompt intervention to prevent proliferative vitreoretinopathy (PVR) and endophthalmitis when serious intraocular damage occurs. The management of OGIs involves initial wound closure within 24 hours, followed by vitrectomy as a secondary surgery. However, there is a lack of consensus regarding the optimal timing of vitrectomy for maximizing visual outcomes. This meta-analysis aimed to investigate whether early or delayed vitrectomy leads to better outcomes in patients with OGIs. Methods: This review was conducted based on PRISMA guidelines. The Medline, Embase, Scopus, Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov databases were searched (October 23, 2023). Clinical studies that used vitrectomy to manage OGIs as early (within 7 days) or delayed (8– 14 days) interventions were included. Randomized controlled trials (RCTs) and non-RCTs were appraised using the Cochrane risk of bias and JBI tools, respectively. Results: Eleven studies met the inclusion criteria and were included in the quantitative analyses. There were 235 patients with OGIs who received early intervention and 211 patients who received delayed intervention. The retina was reattached in 91% and 76% of the patients after early and delayed intervention, respectively. Traumatic PVR was present in 9% and 41% of the patients in the early and delayed groups, respectively. The odds of retinal reattachment after vitrectomy were greater in the early group (OR = 3.42, p = 0.010, 95% CI=1.34– 8.72), and the odds of visual acuity ≥ 5/200 were 2.4 times greater in the early group. The incidence of PVR was significantly greater in the delayed surgery group (OR = 0.16, p < 0.0001; 95% CI=0.06– 0.39), which also required more than one vitrectomy surgery. Conclusion: Early vitrectomy results in better postoperative visual acuity, a greater proportion of retinal reattachment, and a decreased incidence of PVR. [ABSTRACT FROM AUTHOR]

To report a novel protocol for diagnosis of retinal artery occlusions at the point of care using OCT and a remote consult model. Retrospective case series and evaluation of a diagnostic test or technology. Adult patients who presented with painless monocular vision loss and were diagnosed with a nonarteritic retinal artery occlusion. OCT machines were placed in the stroke center or emergency department at 3 hospitals within our health system. Patients who presented with painless monocular vision loss were evaluated by the stroke neurology service and an OCT was acquired. The images were interpreted remotely by the retina service. An in-house ophthalmology consult was not required to make the final treatment decision. Eligible patients were treated with intra-arterial tissue plasminogen activator (IA-tPA). Patients were followed by ophthalmology during their admission when an in-house consultation service was available or otherwise evaluated immediately after discharge. Visual acuity (VA) before and after treatment with IA-tPA; time from last known well (LKW) to treatment; and time from presentation to treatment. In the first 18 months since the protocol went live, 59 patients were evaluated. Twenty-five patients (42%) had a confirmed retinal artery occlusion based on OCT and follow-up examination. Ten patients were eligible for treatment, and 9 patients received treatment with IA-tPA. There was a statistically significant improvement in mean VA from logarithm of the minimum angle of resolution (logMAR) 2.14 to logMAR 0.7 within 24 hours after treatment (P = 0.0001) and logMAR 1.04 after 4 weeks (P = 0.01). Clinically significant improvement was noted in 66% of patients within 24 hours and maintained through 1 month in 56% of all treated patients. The mean time to treatment from LKW was 543 minutes and from presentation at the stroke center was 146 minutes. We report the successful implementation of a remote consult protocol using point-of-care automated OCT. This novel paradigm demonstrates the potential utility of remote consult services for the diagnosis of time-sensitive ophthalmic emergencies. Proprietary or commercial disclosure may be found after the references. [ABSTRACT FROM AUTHOR]

Rana, Vipin, Pradeep Kumar, Sandeepan Bandopadhyay, Vijay K. Sharma, Meenu Dangi, Dattakiran Joshi, Sanjay Kumar Mishra, Satyabrat Srikumar, and V.A. Arun. Central retinal artery occlusion in young adults at high altitude: thin air, high stakes. High Alt Med Biol. 00:000–000, 2024.—We present five cases of young security personnel who were posted at high altitude (HA) for a duration of at least 6 months and presented with a sudden decrease of vision in one eye. The diagnosis of central retinal artery occlusion (CRAO) was made in all patients. Fundus fluorescein angiography and optical coherence tomography of the macula supported the diagnosis. None of these cases had any preexisting comorbidities. Erythrocytosis was noticed in all patients, and two of them had hyperhomocysteinemia. Four out of five patients showed either middle cerebral artery or internal carotid artery (ICA) thrombosis on computed tomography angiography. The patients were managed by a team of ophthalmologist, hematologist, vascular surgeon, and neurologist. In cases of incomplete ICA occlusion, patients were managed surgically. However, in the case of complete ICA occlusion, management was conservative with antiplatelet drugs. This case series highlights HA-associated erythrocytosis and hyperhomocysteinemia as important risk factors for CRAO in young individuals stationed at HA. [ABSTRACT FROM AUTHOR]

Objectives: A real-time and robust 3D dynamic object perception module is a key part of autonomous driving system. Methods: This paper fuses monocular camera and light detection and ranging (LiDAR) to detect 3D objects. First, we use convolutional neural network to detect 2D bounding boxes and generate 3D frustum region of interest (ROI) according to the geometric projection relation between camera and LiDAR. Then, we cluster the point cloud in the frustum ROI and fit the 3D bounding box of the objects. After detecting 3D objects, we reidentify the objects between adjacent frames by appearance features and Hungarian algorithm, and then propose a tracker management model based on a quad-state machine. Finally, a novel prediction model is proposed, which leverages lane lines to constrain vehicle trajectories. Results: The experimental results show that in the stage of target detection, the accuracy and recall of the proposed algorithm can reach 92.5% and 86.7%, respectively. The root mean square error of the proposed trajectory prediction algorithm is smaller than that of the existing algorithms on the simulation datasets including straight line, arc and spiral curves. The whole algorithm only takes approximately 25 ms, which meets the real-time requirements. Conclusions: The proposed algorithm is effective and efficient, and has a good performance in different lane lines. [ABSTRACT FROM AUTHOR]

Transient monocular visual loss (TMVL), also known as transient monocular blindness or amaurosis fugax ("fleeting blindness"), is a temporary loss of vision often due to ischemia to the retina. While acute TMVL should be considered an emergency that further requires exhaustive investigation, there are some cases in which TMVL arises secondary to benign causes. Age has a major impact in the diagnosis of ischemia and although the differential diagnosis of TMVL can be broad, timely and appropriate history, examination, diagnostic testing, and treatment can be vision- or life-saving. We review the causes of TMVL and the impact of age on the differential diagnoses and management. [ABSTRACT FROM AUTHOR]

Rockfall intrusion detection is crucial for the safety management of railway operations, and video detection methods help reduce deployment costs and improve detection efficiency. Mainstream neural network-based video detection methods have rapidly evolved in recent years but struggle to adapt to complex scenarios such as existing railway slope constructions due to weak generalization ability, low accuracy, and limited information acquisition. Therefore, this paper introduces a dynamic neural network detection model and establishes a dataset for rockfall intrusions in existing railway slope scenarios. The model initially relies on the YOLOv5 neural network, adopting an activation function suitable for the target scenario and addressing overfitting to achieve precise target recognition. Based on the neural network, the model dynamically detects rolling rockfalls by integrating a background subtraction algorithm based on the Gaussian Mixture Model and captures target dimensions using monocular vision technology, thus broadening the dimensions of detection information. Trials conducted on a railway in Shandong, China, demonstrate that the model accurately identifies moving rockfalls along the railway slopes and acquires the dimensions of moving rockfalls, successfully filtering out low-risk targets in the scene. [ABSTRACT FROM AUTHOR]

Abstract As industries develop, the automation and intelligence level of power plants is constantly improving, and the application of patrol robots is also increasingly widespread. This research combines computer vision technology and particle swarm optimization algorithm to build an obstacle recognition model and obstacle avoidance model of an intelligent patrol robot in a power plant respectively. Firstly, the traditional convolutional recurrent neural network is optimized, and the obstacle recognition model of an intelligent patrol robot is built by combining the connection timing classification algorithm. Then, the artificial potential field method optimizes the traditional particle swarm optimization algorithm, and an obstacle avoidance model of an intelligent patrol robot is built. The performance of the two models was tested, and it was found that the highest precision, recall, and F1 values of the identification model were 0.978, 0.974, and 0.975. The highest precision, recall, and F1 values of the obstacle avoidance model were 0.97, 0.96, and 0.96 respectively. The two optimization models designed in this research have better performance. In conclusion, the two models in this study are superior to the traditional methods in recognition effect and obstacle avoidance efficiency, providing an effective technical scheme for intelligent patrol inspection of power plants.

Objective: The study aimed to evaluate the influence of ocular dominance, clinical experience, and sex on the accuracy of visual color matching. Method and materials: A total of 104 selected participants consisting of dental students and dental practitioners were allocated into two groups based on their clinical experience of using visual shade matching tabs for the selection of tooth shade. Both the inexperienced (IXP), as well as experienced participants (EXP), were then asked to correctly match the shades of five tabs (A2, A3.5, B1, C2, and D3) of a commonly used visual shade guide (Vitapan Classical shade guide, VITA Zahnfabrik) that were kept constant throughout the study. The participants documented the shade of those tabs using another set of shade guide using the right eye (left eye covered), left eye (right eye covered), as well as both eyes. Test for ocular dominance was performed using the Dolman method. The results were statistically analyzed using the chi-square test with a level of signifi- cance set at < .05. Results: The percentage of correct shade matching with the monocular dominant vision (53%) was signifi- cantly better as compared to the monocular non- dominant vision (12%) (P < .001) or binocular vision (44%) (P < .05). Sex had an influence in shade selection (P = .031). However, clinical experience was an insignificant aspect in shade matching (P = .078). Conclusions: Knowing the clinician's dominant eye can be clinically relevant, as it can influence the accuracy of shade selection, irrespective of sex and clinical experience. [ABSTRACT FROM AUTHOR]

Background: Retinal artery occlusions lead to sudden, painless vision loss, affecting millions globally. Despite their significance, treatment strategies remain unestablished, contrasting with acute ischemic stroke (AIS), where IVT has proven efficacy. Similar to AIS, retinal artery occlusions demand urgent evaluation and treatment, reflecting the principle "time is retina". Even for patients with transient monocular vision loss, also known as amaurosis fugax (AF), pertinent guidelines meanwhile recommend immediate emergency assessment in a specialized facility. However, data on the clinical benefit and comparability with persistent occlusions are missing. This study aimed to compare the results of a comprehensive stroke-workup among patients with persistent retinal artery occlusions (RAO), including both central retinal (CRAO) and branch retinal artery occlusion (BRAO) and those with AF. Methods: Conducted at the University Hospital Giessen, Germany, this exploratory cross-sectional study enrolled patients with transient or permanent unilateral vision loss of non-arteritic origin. The primary outcome were differences between the two groups RAO and AF with regard to cardiovascular risk profiles and comorbidities, vascular and pharmacological interventions and clinical neurological and ophthalmological outcomes. Secondary outcome was a sub-group analysis of patients receiving IVT. Results: Out of 166 patients assessed, 76 with RAO and 40 with AF met the inclusion criteria. Both groups exhibited comparable age, gender distribution, and cardiovascular risk profiles. Notably, RAO patients did not show significantly more severe vascular comorbidities than AF patients. However, AF patients received vascular interventions more frequently. Pharmacological intervention rates were similar across groups. RAO patients had slightly worse neurological outcomes, and IVT did not yield favorable ophthalmological outcomes within any observed patients. Conclusion: The study found similar vascular burden and risk factors in patients with RAO and AF, with implications for clinical workflows. IVT for RAO may only be effective in very early treatment windows. This emphasizes the need for public awareness and collaborative protocols between ophthalmologists and neurologists to improve outcomes. [ABSTRACT FROM AUTHOR]

This study aims to examine whether binocular vision plays a facilitating or impeding role in lexical processing during sentence reading in Chinese. Adopting the revised boundary paradigm, we orthogonally manipulated the parafoveal and foveal viewing conditions (monocular vs. binocular) of target words (high- vs. low-frequency) within sentences. Forty participants (30 females, mean age = 19.9 years) were recruited to read these sentences and their eye movements were monitored. Through directly comparing the eye movement measures in different viewing conditions, the results indicated that compared with monocular viewing, binocular viewing resulted in shorter fixation durations, thereby facilitating lexical processing. Critically, in addition to the higher information encoding speed toward the currently fixated word in the fovea, the more efficient preprocessing of the upcoming text to the right of fixation in the parafovea may also contribute to the superiority of binocular vision over monocular. Our findings provide the first evidence to support the binocular advantages in Chinese reading, which reveals that high-quality visual input from binocular vision plays a vital role in fluent and efficient written text reading. [ABSTRACT FROM AUTHOR]

This paper presents PAL-SLAM2, a visual and visual–inertial monocular simultaneous localization and mapping (SLAM) system for a panoramic annular lens (PAL) with an ultra-hemispherical field of view (FoV), overcoming the limitations of traditional frameworks in handling fast turns, nighttime conditions and rapid lighting changes. The system incorporates modules for initialization, tracking, local mapping, loop and map merging. To fully exploit information from the negative space (z < 0), keypoints are projected onto the unit sphere for visual initialization and tightly-coupled visual–inertial optimization. Leveraging a multimap strategy, the feature-based PAL-SLAM2 is capable of detecting unidirectional and bidirectional public areas using PAL images, thereby enhancing the performance of loop correction and map merging. Testing indicates that it achieves an average accuracy of 7.1 cm on the PALVIO indoor dataset, surpassing similar state-of-the-art frameworks. Furthermore, we have collected a large-scale dataset comprising over 120,000 images with corresponding inertial measurement unit (IMU) data, demonstrating PAL-SLAM2's robustness under challenging outdoor conditions. Relying solely on visual and inertial inputs, our system is particularly suited for environments where Global Navigation Satellite System (GNSS) signals are frequently obstructed, such as indoor spaces or dense urban areas. The dataset can be accessed at: https://github.com/wwendy233/Plaza-Dataset. [ABSTRACT FROM AUTHOR]

Super multi-view (SMV) display is a promising 3D display technology; however, potential frame loss due to bandwidth-limited video transmission could cause discomfort to viewers. Thus, an evaluation of the acceptable viewing experience will be valuable. This study investigates the effects of frame loss on visual fatigue in SMV display, focusing on quantified frame loss rates and varying frame loss modes. Experiments were conducted with 20 subjects, utilizing the Stroop test through an SMV display system to evaluate the visual fatigue under different frame loss conditions. The results show a rise in visual fatigue as the frame loss rate increases, with two critical thresholds identified. A 4% frame loss rate marks the threshold for significant loss-induced differences, beyond which visual fatigue begins to become significant in loss-induced modes compared to the normal loss-free mode. Subsequently, a 10% frame loss rate marks the threshold for significant mode-dependent differences, beyond which variations appear between different loss-induced modes, with monocular mode inducing more visual fatigue than binocular and dual-view more than single-view. Consequently, the findings advocate for refining the 3D video processing to maintain a frame loss rate below 4% for negligible effect and considering the interactions between different views for less visual fatigue. This research aims to provide insights and guidance for addressing potential challenges in developing and applying SMV display technology. [ABSTRACT FROM AUTHOR]

Monocular stereo vision has excellent application prospects in the field of microrobots. On the basis of the geometric model of bifocal imaging, this paper proposes a monocular depth perception method by liquid zoom imaging. Firstly, the configuration of a monocular liquid vision system for depth measurement is presented, and the working mechanism of the system is analyzed through theoretical derivation. Then, to eliminate the influence of optical axis drift induced by the liquid gravity factor on the measurement results, the target image area is used as the calculation feature instead of the image vector length. A target area calculation method based on chain code classification and strip segmentation is proposed. Furthermore, in response to the fluctuation problem of liquid lens focal power caused by factors such as temperature and object distance, a dynamic focal length model of the liquid zoom imaging system is constructed after precise calibration of the focal power function. Finally, a testing experiment is designed to validate the proposed method. The experimental results show that the average error of depth perception methods is 4.30%, and its measurement time is only on the millisecond scale. Meanwhile, the proposed method has good generalization performance. [ABSTRACT FROM AUTHOR]

To address the issues of poor detection accuracy and the large number of target detection model parameters in existing AGV monocular vision location detection algorithms, this paper presents an AGV vision location method based on Gaussian saliency heuristic. The proposed method introduces a fast and accurate AGV visual detection network called GAGV-net. In the GAGV-net network, a Gaussian saliency feature extraction module is designed to enhance the network's feature extraction capability, thereby reducing the required output for model fitting. To improve the accuracy of target detection, a joint multi-scale classification and detection task header are designed at the stage of target frame regression to classification. This header utilizes target features of different scales, thereby enhancing the accuracy of target detection. Experimental results demonstrate a 12% improvement in detection accuracy and a 27.38 FPS increase in detection speed compared to existing detection methods. Moreover, the proposed detection network significantly reduces the model's size, enhances the network model's deployability on AGVs, and greatly improves detection accuracy. [ABSTRACT FROM AUTHOR]

This survey reviews advances in 3D object detection approaches for autonomous driving. A brief introduction to 2D object detection is first discussed and drawbacks of the existing methodologies are identified for highly dynamic environments. Subsequently, this paper reviews the state-of-the-art 3D object detection techniques that utilizes monocular and stereo vision for reliable detection in urban settings. Based on depth inference basis, learning schemes, and internal representation, this work presents a method taxonomy of three classes: model-based and geometrically constrained approaches, end-to-end learning methodologies, and hybrid methods. There is highlighted segment for current trend of multi-view detectors as end-to-end methods due to their boosted robustness. Detectors from the last two kinds were specially selected to exploit the autonomous driving context in terms of geometry, scene content and instances distribution. To prove the effectiveness of each method, 3D object detection datasets for autonomous vehicles are described with their unique features, e. g., varying weather conditions, multi-modality, multi camera perspective and their respective metrics associated to different difficulty categories. In addition, we included multi-modal visual datasets, i. e., V2X that may tackle the problems of single-view occlusion. Finally, the current research trends in object detection are summarized, followed by a discussion on possible scope for future research in this domain. [ABSTRACT FROM AUTHOR]

When traversing soft and rugged terrain, a planetary rover is susceptible to slipping and sinking, which impedes its movement. The real-time detection of wheel sinkage in the planetary wheel-on-limb system is crucial for enhancing motion safety and passability on such terrain. Initially, this study establishes a measurement of wheel sinkage under complex terrain conditions. Subsequently, a monocular vision-based wheel sinkage detection method is presented by combining the wheel–terrain boundary with the wheel center position (WTB-WCP). The method enables the efficient and accurate detection of wheel sinkage through two-stage parallel computation of the wheel–terrain boundary fitting and wheel center localization. Finally, this study establishes an experimental platform based on a monocular camera and the planetary rover wheel-on-limb system to experimentally validate and comparatively analyze the proposed method. The experimental results demonstrate that the method effectively provides information on the wheel sinkage of the planetary rover wheel-on-limb system, and the relative errors of the method do not exceed 4%. The method has high accuracy and reliability and is greatly significant for the safety and passability of planetary rovers in soft and rugged terrain. [ABSTRACT FROM AUTHOR]

Introduction: Changes in vision that occur in normal healthy aging can be seen in fundamental measures of monocular vision. However, the nature of the changes in binocular vision with age remain unclear. Methods: A total of 28 older (53-66 years) and 28 younger adults (20-31 years) were enrolled in this study. We performed a battery of tests to assess differences in monocular contrast thresholds and various binocular visual functions including dichoptic masking weight and strength, the binocular balance point for fused stimuli, and stereoacuity in the aging and control groups. Results: Aging significantly increased monocular contrast thresholds (p < 0.001). Although this suggests that aging reduces the effective "input gain" to vision, we also found a significantly elevated contribution of those weaker signals to interocular suppression (p < 0.001). Consequently, there was no significant net difference in the strength of interocular suppression (p = 0.065). We did not find a significant difference of absolute balance point between the two groups (p = 0.090). Lastly, the mean stereoacuity was worse in the older group compared to the younger group (p = 0.002). Discussion: Our findings confirm previous results showing differences in contrast sensitivity and stereoacuity with aging. Furthermore, we find a change in interocular suppression that is a possible consequence of the change in contrast sensitivity. It is suggestive of a cortical system that maintains a homeostatic balance in interocular suppression across the lifespan. [ABSTRACT FROM AUTHOR]

Background. The coronavirus disease (COVID-19) is a highly contagious disease with profound health implications. It can affect any part of the body with variable severity. Various ophthalmic manifestations of coronavirus disease have been documented. Case Presentations. We reported three cases of outer retinopathies associated with COVID-19 infection. All three patients were young females. The first two patients presented within days of COVID-19 infection with complaints of black spots in the eyes. Multimodal retinal imaging showed lesions consistent with acute macular neuroretinopathy. Lesions were bilateral in the first patient and unilateral in the second one. Our third patient presented with blurred vision in one eye, 3 months after a suspected COVID-19 infection. Retinal imaging showed outer retinopathy. Our patients' vision was good and maintained during the follow-up. All three were monitored on observation only, and symptoms and lesions improved with time. Conclusion. In conclusion, COVID-19-related thromboinflammatory response can result in localized vascular inflammation and hypoperfusion in any of the retinal capillary plexuses or choriocapillaris resulting in ischemia of the corresponding retinal or choroidal layers. [ABSTRACT FROM AUTHOR]

This document explores an alternative method for achieving spectacle independence in the treatment of presbyopia. Traditional approaches have limitations, such as reduced visual quality or stereoacuity. Researchers have developed new intraocular lenses and corneal procedures that create an extended depth of focus, allowing for continuous vision at different distances. Two specific approaches, Laser Blended Vision and bilateral implantation of extended depth of focus intraocular lenses, have shown promising results in achieving spectacle independence. The document presents case studies of patients who underwent implantation of the extended depth of focus intraocular lens and achieved clear vision without glasses. Overall, this strategy offers an alternative option for patients who are not suitable candidates for laser blended vision therapy. [Extracted from the article]

Urban visual localization is the process of determining the pose (position and attitude) of the imaging sensor (or platform) with the help of existing geo-referenced data. This task is critical and challenging for many applications, such as autonomous navigation, virtual and augmented reality, and robotics, due to the dynamic and complex nature of urban environments that may obstruct Global Navigation Satellite Systems (GNSS) signals. This paper proposes a block-wise matching strategy for urban visual localization by using geo-referenced Google Street View (GSV) panoramas as the database. To determine the pose of the monocular query images collected from a moving vehicle, neighboring GSVs should be found to establish the correspondence through image-wise and block-wise matching. First, each query image is semantically segmented and a template containing all permanent objects is generated. The template is then utilized in conjunction with a template matching approach to identify the corresponding patch from each GSV image within the database. Through the conversion of the query template and corresponding GSV patch into feature vectors, their image-wise similarity is computed pairwise. To ensure reliable matching, the query images are temporally grouped into query blocks, while the GSV images are spatially organized into GSV blocks. By using the previously computed image-wise similarities, we calculate a block-wise similarity for each query block with respect to every GSV block. A query block and its corresponding GSV blocks of top-ranked similarities are then input into a photogrammetric triangulation or structure from motion process to determine the pose of every image in the query block. A total of three datasets, consisting of two public ones and one newly collected on the Purdue campus, are utilized to demonstrate the performance of the proposed method. It is shown it can achieve a meter-level positioning accuracy and is robust to changes in acquisition conditions, such as image resolution, scene complexity, and the time of day. [ABSTRACT FROM AUTHOR]

In the field of medical applications, precise localization of medical instruments and bone structures is crucial to ensure computer-assisted surgical interventions. In orthopedic surgery, existing devices typically rely on stereoscopic vision. Their purpose is to aid the surgeon in screw fixation of prostheses or bone removal. This article addresses the challenge of localizing a rigid object consisting of randomly arranged planar markers using a single camera. This approach is especially vital in medical situations where accurate object alignment relative to a camera is necessary at distances ranging from 80 cm to 120 cm. In addition, the size limitation of a few tens of centimeters ensures that the resulting locator does not obstruct the work area. This rigid locator consists of a solid at the surface of which a set of plane markers (ArUco) are glued. These plane markers are randomly distributed over the surface in order to systematically have a minimum of two visible markers whatever the orientation of the locator. The calibration of the locator involves finding the relative positions between the individual planar elements and is based on a bundle adjustment approach. One of the main and known difficulties associated with planar markers is the problem of pose ambiguity. To solve this problem, our method lies in the formulation of an efficient initial solution for the optimization step. After the calibration step, the reached positioning uncertainties of the locator are better than two-tenth of a cubic millimeter and one-tenth of a degree, regardless of the orientation of the locator in space. To assess the proposed method, the locator is rigidly attached to a stylus of about twenty centimeters length. Thanks to this approach, the tip of this stylus seen by a 16.1 megapixel camera at a distance of about 1 m is localized in real time in a cube lower than 1 mm side. A surface registration application is proposed by using the stylus on an artificial scapula. [ABSTRACT FROM AUTHOR]

Vision-based autonomous underwater vehicle localization, navigation, and mapping is an attractive field of research in recent years. Despite the development of vision-only and vision-inertial-based algorithms, particularly in the terrestrial and aerial robotics groups, applying such approaches directly to underwater environments remains challenging due to visual degradation in the medium and further due to the slow crusade of the underwater vehicles. This paper introduces U-VIP-SLAM, a unique visual-inertial-pressure SLAM system that combines data from a monocular camera, a low-cost IMU, and a pressure sensor to create a SLAM system focused to the underwater region. All of the key components of a comprehensive SLAM system are included in the U-VIP-SLAM, including loop closure capabilities tailored to the underwater environment. Further, we introduce forward and backward projection strategies to fuse pressure measurement in the local bundle adjustment. Furthermore, a unique initialization strategy is proposed to effectively bootstrap the system by fusing pressure and IMU data. Several datasets were used to compare the performance of the proposed methods against the state-of-the-art monocular VI systems. Source Code is available at https://github.com/chintha/U-VIP-SLAM.git. [ABSTRACT FROM AUTHOR]

The lack of discernible vehicle contour features in low-light conditions poses a formidable challenge for nighttime vehicle detection under hardware cost constraints. Addressing this issue, an enhanced histogram of oriented gradients (HOGs) approach is introduced to extract relevant vehicle features. Initially, vehicle lights are extracted using a combination of background illumination removal and a saliency model. Subsequently, these lights are integrated with a template-based approach to delineate regions containing potential vehicles. In the next step, the fusion of superpixel and HOG (S-HOG) features within these regions is performed, and the support vector machine (SVM) is employed for classification. A non-maximum suppression (NMS) method is applied to eliminate overlapping areas, incorporating the fusion of vertical histograms of symmetrical features of oriented gradients (V-HOGs). Finally, the Kalman filter is utilized for tracking candidate vehicles over time. Experimental results demonstrate a significant improvement in the accuracy of vehicle recognition in nighttime scenarios with the proposed method. [ABSTRACT FROM AUTHOR]

Aiming at the shortcomings of the existing machine vision pose measurement technology, a pose measurement method based on monocular vision and a cooperative target is proposed. A planar target designed with circles and rings as the main body is dedicated to object pose measurement, and a feature point coordinate extraction and sorting algorithm is designed for this target to effectively extract image features on the target. The RANSAC algorithm and topology-based fitting for the intersection method are used to optimise data processing, further improving the accuracy of feature point coordinate extraction and ultimately achieving high-precision measurement of object poses. The experimental results show that the measurement accuracy of the roll angle perpendicular to the optical axis can reach 0.02°, and the repeatability can reach 0.0004° after removing the systematic error; the measurement accuracy of the pitch angle can reach 0.03°, and the repeatability can go to 0.002° after removing the systematic error. The measurement range of the pitch angle is [−30°, +30°]; the measurement range of the roll angle is [−179°, +179°]. The experimental results show that the system has high measurement accuracy and meets the requirements of high-precision measurement. [ABSTRACT FROM AUTHOR]

Optic perineuritis is the inflammation of the optic nerve sheath. This affliction can lead to visual field impairment and other signs and symptoms related to the orbital space, such as pain, disc edema, ophthalmoplegia, proptosis. However, not all patients present with such suggestive symptoms, requiring a thorough assessment. We report the case of a young male admitted to our hospital for recurrent episodes of monocular blindness. Amaurosis fugax is a well-known presentation of transient ischemic attacks (TIA) and it was ruled out. Gadolinium-enhanced MRI revealed a typical aspect of optic perineuritis. It was mandatory to consider all possible causes of secondary optic perineuritis as they all represent serious clinical conditions, even if the idiopathic form is more frequent. The clinical and paraclinical evaluation of the patient excluded an underlying disease and primary optic perineuritis was diagnosed. Corticosteroid therapy is usually curative and a course of methylprednisolone was initiated for our patient with good outcome. However, response to treatment is not diagnostic as both primary and secondary optic perineuritis are normally responsive, hence thorough differential diagnosis is necessary. [ABSTRACT FROM AUTHOR]

Enhancing the autonomous landing capability of unmanned aerial vehicles (UAVs) is of great significance for improving their operational efficiency and field survival capabilities. To this end, we propose a real-time autonomous landing spot detection method for UAVs. Firstly, the pose of the UAV at any given time and the initial three-dimensional point cloud of the scene are estimated using simultaneous localization and mapping (SLAM) techniques. Then, since the initial point cloud is sparse and cannot be used for terrain analysis, we generate a voxel-based elevation map of the scene, which can establish interconnectivity among the points. Finally, we propose a shift-box strategy to comprehensively analyze various terrain factors in the elevation map, determine the landing spot for UAVs, and update the landing spot in real time. The UAV flight experiments conducted in the real world have demonstrated the effectiveness and real-time performance of the proposed method. [ABSTRACT FROM AUTHOR]

ABSTRACT In order to realise automatic row alignment for sugar beet harvesting, and to improve the efficiency of sugar beet combine harvesters, it is essential to be able to detect the positions of sugar beet rows. In this paper, we consider beet harvesting as a scenario, and an extraction algorithm for beet crop rows is proposed based on videos of beets collected with a monocular camera. We focus on the characteristics of the root and stem features, which are relatively difficult to distinguish from the surrounding environment after cutting the tops and beating the leaves of sugar beet, and a crop row extraction algorithm is designed in which threshold segmentation is carried out based on histogram equalisation and MM-Filter (median filtering and minimum area filtering) is applied to remove noise. The proposed algorithm retains more complete feature information and has better denoising effect than the traditional algorithm, and the processing time is only 11.751 ms, a value 6.9% lower than for the alternative approach. Our algorithm is feasible in terms of recognition and extraction, and can provide a theoretical basis for the construction of an automatic row-to-row control system for a sugar beet combine harvester. It can also provide technical support for the design of an automatic sugar beet combine harvester.

Three-dimensional object detection has been substantially improved with the use of expensive LiDAR and stereo vision systems in intelligent driving. The less-expensive and more scalable solution of monocular 3D object detection, however, remains a key challenge. This study primarily explores real-time pseudo 3D object detection with monocular vision and designs a single-shot RPN model, VKP-P3D, which relies purely on visual feature extraction. Through a multiscale feature fusion and an attention mechanism module, this model obtains high-dimensional feature representations during the feature extraction phase. In the detection head of the VKP-P3D model, the pseudo 3D object detection is obtained by regressing 2D bounding box and the visible key points within the image coordinate of the 3D box from the camera’s perspective. Finally, assuming flat ground and considering geometric parameters of the camera, the object’s 3D information can be extracted. To verify the effectiveness of the proposed algorithm, we constructed two pseudo 3D object detection datasets based on visible key points and compared with current state-of-the-art real-time object detector. Results showed that the proposed model has high detection accuracy and speed.

Monocular 3D object tracking is a challenging task because monocular image lacks depth information necessary for 3D scene understanding. Modern methods typically rely on deep learning to reconstruct 3D information from learned prior, which demands strenuous effort on acquiring ground-truth annotated data and does not generalize for various camera settings. We present a method to continuously track 3D location and orientation of the target object from a monocular image sequence from 2D instance segmentation methods. We reconstruct the structure and trajectory of the objects using factor graph optimization incorporating reprojection error of keypoint tracks, kinematic motion model and bounding box constraints. We propose a combined ellipsoid-cuboid object representation and bounding box constraint to model the object dimension. We evaluate our algorithm in simulation dataset generated using CARLA, and the result indicates that the method is robust to 2D bounding box error and the proposed object representation yields more accurate pose and size estimation compared to solely using either representation.