This gives some information about the conformal point and the calibrating conic, and their relationship one to the other. These concepts are useful for visualizing image geometry, and lead to intuitive ways to compute geometry, such as angles and directions in an image.
Consumer-grade camera systems often struggle to maintain stable image quality under complex illumination conditions such as low light, high dynamic range, and backlighting, as well as spatial color temperature variation. These issues lead to underexposure, color casts, and tonal inconsistency, which degrade the performance of downstream vision tasks. To address this, we propose ACamera-Net, a lightweight and scene-adaptive camera parameter adjustment network that directly predicts optimal exposure and white balance from RAW inputs. The framework consists of two modules: ACamera-Exposure, which estimates ISO to alleviate underexposure and contrast loss, and ACamera-Color, which predicts correlated color temperature and gain factors for improved color consistency. Optimized for real-time inference on edge devices, ACamera-Net can be seamlessly integrated into imaging pipelines. Trained on diverse real-world data with annotated references, the model generalizes well across lighting conditions. Extensive experiments demonstrate that ACamera-Net consistently enhances image quality and stabilizes perception outputs, outperforming conventional auto modes and lightweight baselines without relying on additional image enhancement modules.
Detecting hate speech in videos remains challenging due to the complexity of multimodal content and the lack of fine-grained annotations in existing datasets. We present HateClipSeg, a large-scale multimodal dataset with both video-level and segment-level annotations, comprising over 11,714 segments labeled as Normal or across five Offensive categories: Hateful, Insulting, Sexual, Violence, Self-Harm, along with explicit target victim labels. Our three-stage annotation process yields high inter-annotator agreement (Krippendorff's alpha = 0.817). We propose three tasks to benchmark performance: (1) Trimmed Hateful Video Classification, (2) Temporal Hateful Video Localization, and (3) Online Hateful Video Classification. Results highlight substantial gaps in current models, emphasizing the need for more sophisticated multimodal and temporally aware approaches. The HateClipSeg dataset are publicly available at https://github.com/Social-AI-Studio/HateClipSeg.git.
In April 2023, the House of Bishops of the Province of the Church of Uganda elected Canon Godfrey Kasana as Bishop of Luwero. Before his consecration could take place, however, a member of the church submitted a petition alleging that he was unsuitable for consecration on grounds of adultery – and in June the House of Bishops revoked his nomination. The respondents, in effect, sought judicial review of that decision, while the Archbishop argued that the claim was brought against the wrong party and was frivolous, vexatious and an abuse of process.
Samuel Gibbon, Graciela Muniz-Terrera, Fabian SL Yii
et al.
Purpose: To develop an automatic method of quantifying optic disc pallor in fundus photographs and determine associations with peripapillary retinal nerve fibre layer (pRNFL) thickness. Methods: We used deep learning to segment the optic disc, fovea, and vessels in fundus photographs, and measured pallor. We assessed the relationship between pallor and pRNFL thickness derived from optical coherence tomography scans in 118 participants. Separately, we used images diagnosed by clinical inspection as pale (N=45) and assessed how measurements compared to healthy controls (N=46). We also developed automatic rejection thresholds, and tested the software for robustness to camera type, image format, and resolution. Results: We developed software that automatically quantified disc pallor across several zones in fundus photographs. Pallor was associated with pRNFL thickness globally (\b{eta} = -9.81 (SE = 3.16), p < 0.05), in the temporal inferior zone (\b{eta} = -29.78 (SE = 8.32), p < 0.01), with the nasal/temporal ratio (\b{eta} = 0.88 (SE = 0.34), p < 0.05), and in the whole disc (\b{eta} = -8.22 (SE = 2.92), p < 0.05). Furthermore, pallor was significantly higher in the patient group. Lastly, we demonstrate the analysis to be robust to camera type, image format, and resolution. Conclusions: We developed software that automatically locates and quantifies disc pallor in fundus photographs and found associations between pallor measurements and pRNFL thickness. Translational relevance: We think our method will be useful for the identification, monitoring and progression of diseases characterized by disc pallor/optic atrophy, including glaucoma, compression, and potentially in neurodegenerative disorders.
Adam Threlfall, Samuel Gibbon, James Cameron
et al.
Background and Objective: Infra-red scanning laser ophthalmoscope (IRSLO) images are akin to colour fundus photographs in displaying the posterior pole and retinal vasculature fine detail. While there are many trained networks readily available for retinal vessel segmentation in colour fundus photographs, none cater to IRSLO images. Accordingly, we aimed to develop (and release as open source) a vessel segmentation algorithm tailored specifically to IRSLO images. Materials and Methods: We used 23 expertly annotated IRSLO images from the RAVIR dataset, combined with 7 additional images annotated in-house. We trained a U-Net (convolutional neural network) to label pixels as 'vessel' or 'background'. Results: On an unseen test set (4 images), our model achieved an AUC of 0.981, and an AUPRC of 0.815. Upon thresholding, it achieved a sensitivity of 0.844, a specificity of 0.983, and an F1 score of 0.857. Conclusion: We have made our automatic segmentation algorithm publicly available and easy to use. Researchers can use the generated vessel maps to compute metrics such as fractal dimension and vessel density.
Abstract Funding Acknowledgements Type of funding sources: None. Introduction A possible side-effect of various medical drugs is prolongation of the electric repolarization of the heart, measured as the corrected QT-interval (QTc). Patients treated with these drugs should be monitored frequently via an ECG to screen for early changes indicating possible life-threating arrythmias. Especially during the Covid-19 pandemic, remote patient monitoring gained importance. The Withings Scanwatch offers automated analysis of the QTc remotely, thereby obviating the need for in-person visits. We aimed to compare automated QTc-measurements using a single lead ECG (SL-ECG) of a novel smartwatch (Withings Scanwatch, SW-ECG) with manual-measured QTc from a nearly simultaneously recorded standard 12-lead ECG. Methods We enrolled consecutive patients referred to a tertiary hospital for cardiac workup in a prospective, observational study. To obtain a SW-ECG, patients were instructed to keep their index finger on the stainless steel ring on the top case of the smartwatch continuously for 30 seconds The QT-interval was manually interpreted by two blinded, independent cardiologists through the tangent-method, using lead II or V5/V6. Bazett’s formula was used to calculate QTc. Results We prospectively enrolled 317 patients (48% female, mean age 63.3 ± 17.2 years). The smartwatch was able to automatically measure QTc-intervals in 177 patients (56%). The diagnostic accuracy of SW-ECG for detection of a QTc-interval ≥ 460ms as quantified by the area under the curve (AUC) was 0.91 (95%CI 86.4-95.9). The Bland-Altman analysis resulted in a bias of 6.6ms (95% limit of agreement (LoA) –58.6ms to 71.9ms) comparing automated QTc measurements via SW-ECG with manual QTc-measurement via 12-lead ECG (Figure 1). In 12 patients (6.9%) the difference between the two measurements was greater than the LoA. Premature ventricular complexes, noise or differences in heart rate were responsible in 8.3%, 83.0% and 8.3%, respectively, for observed outliers. Conclusion In this clinical validation of a direct-to-consumer smartwatch we found fair to good agreement between automated-SW-ECG QTc-measurements and manual 12-lead-QTc measurements. The SW-ECG, however, was only able to automatically calculate QTc-intervals in one half of all assessed patients. Our work shows, that the automated algorithm of the SW-ECG needs to be improved to be useful in a clinical setting.
We present a conceptually simple and intuitive method to calculate and to measure the dissimilarities among 2D shapes. Several methods to interpret and to visualize the resulting dissimilarity matrix are presented and compared.
How to fully utilize polarization to enhance synthetic aperture radar (SAR) ship classification remains an unresolved issue. Thus, we propose a dual-polarization information guided network (DPIG-Net) to solve it.
Felix Marattukalam, Waleed H. Abdulla, Akshya Swain
The use of deep learning methods to extract vascular biometric patterns from the palm surface has been of interest among researchers in recent years. In many biometric recognition tasks, there is a limit in the number of training samples. This is because of limited vein biometric databases being available for research. This restricts the application of deep learning methods to design algorithms that can effectively identify or authenticate people for vein recognition. This paper proposes an architecture using Siamese neural network structure for few shot palm vein identification. The proposed network uses images from both the palms and consists of two sub-nets that share weights to identify a person. The architecture performance was tested on the HK PolyU multi spectral palm vein database with limited samples. The results suggest that the method is effective since it has 91.9% precision, 91.1% recall, 92.2% specificity, 91.5%, F1-Score, and 90.5% accuracy values.
This brief paper reports the shortcoming of a variant of convolutional neural network whose components are developed based on the pattern theory framework.
Natalia Díaz-Rodríguez, Vincenzo Lomonaco, David Filliat
et al.
Continual learning consists of algorithms that learn from a stream of data/tasks continuously and adaptively thought time, enabling the incremental development of ever more complex knowledge and skills. The lack of consensus in evaluating continual learning algorithms and the almost exclusive focus on forgetting motivate us to propose a more comprehensive set of implementation independent metrics accounting for several factors we believe have practical implications worth considering in the deployment of real AI systems that learn continually: accuracy or performance over time, backward and forward knowledge transfer, memory overhead as well as computational efficiency. Drawing inspiration from the standard Multi-Attribute Value Theory (MAVT) we further propose to fuse these metrics into a single score for ranking purposes and we evaluate our proposal with five continual learning strategies on the iCIFAR-100 continual learning benchmark.
A field experiment was carried out at the Experimental Farm, Department of Horticulture, School of Agricultural Sciences and Rural Development, Nagaland University, Medziphema Campus, during 2012-2013 and 2013-2014 to find out the response of integrated nutrient management on growth, yield and quality of broccoli cv. Calabrese under the foothill condition of Nagaland. Results revealed that integrated application of 50% NPK + 50% vermicompost + Biofertilizers recorded significantly higher plant height (70.33 cm), number of leaves (20.45), stem diameter (3.25 cm), plant spread (74.25 cm), equilateral head diameter (16.97 cm), polar head diameter(18.37 cm), head size (310.59 cm2), gross head weight (956.0 g), net head weight (330.86 g), yield ha-1 (11.96 t ha-1), protein content (3.77%), vitamin C content (132.50 mg 100 g-1), total soluble solids (5.52) and pH of the head (5.12). While treatment T16- 50% NPK + 50% Pig manure + Biofertilizers gave the highest net return (¹ 260450) and cost benefit ratio (1:3.48) and was found to be significantly at par for all growth, yield and quality parameters with treatment T18 (50% NPK + 50% vermicompost + biofertilizers).
This article provides a step by step development of designing a Object Detection scheme using the HOG based Feature Pyramid aligned with the concept of Template Matching.
Most feature detectors such as edge detectors or circle finders are statistical, in the sense that they decide at each point in an image about the presence of a feature, this paper describes the use of Bayesian feature detectors.
It is proposed a complex valued channel encoding for multidimensional data. The basic approach contains overlapping of complex nonlinear mappings. Its development leads to sparse representation of multi-channel data, increasing their dimensions and the distance between the images.