Uncover the latest and most impactful research in Intervision. Explore pioneering discoveries, insightful ideas and new methods from leading researchers in the field.
Human pose estimation plays a fundamental role in understanding human behaviors and interactions in complex scenes. Among existing approaches, High-Resolution Networks (HRNet) and their variants have demonstrated strong performance in dense...
With the rapid growth of multimodal content on social media, multimodal aspect-based sentiment analysis (MABSA) has become important for fine-grained opinion understanding. Existing methods have attempted to incorporate visual information to enhance...
All-in-one image restoration aims to recover high-quality images from diverse and unknown degradations using a unified model. Existing degradation-aware methods mainly exploit degradation information to modulate latent features, prompts, or...
This work presents an automated, field-implemented framework that links an anomaly detector (PatchCore with a ResNet50 backbone) to a Vision-Language Model (VLM) to produce concise, inspection-ready reports from imagery. The approach is demonstrated...
Minority and underserved youth are less likely to undergo diabetic retinopathy screening, yet more likely to have diabetic retinopathy. We hypothesize that autonomous artificial intelligence-based diabetic eye exams could mitigate screening...
Pose estimation is fundamental to visual simultaneous localization and mapping (SLAM) and 3D perception, yet conventional methods can be computationally demanding and less effective in dynamic multi-view environments. This paper presents...
Mainstream appearance-based gaze estimation methods trained on centralized data are struggling in two urgent aspects. First, most gaze estimation methods require entire eye images or even full faces for model training, which raises serious concerns...
Background Glaucoma is an irreversible blinding disease for which early detection is essential. Precise segmentation of the optic disc (OD) and optic cup (OC) enables accurate cup-to-disc ratio (CDR) calculation, which serves as an important...
Introduction Current AI models for optical coherence tomography (OCT) primarily focus on diagnosis as a standalone task. We developed and validated an end-to-end agentic AI workflow that integrates automated macular hole (MH) diagnosis from OCT...
Automatic short-video editing involves several related tasks, including action recognition, caption generation, frame trimming, transition generation, and color grading. Existing approaches commonly address only part of this process, making it...
Background Minimally invasive surgery (MIS) videos support education and quality review, but raw recordings contain non-informative frames that impede review and increase storage. Methods We retrospectively analyzed 300,832 frames from 120 MIS videos...
Photovoltaic (PV) modules are widely used in building-integrated photovoltaic systems and renewable energy infrastructure. Accurate defect detection is essential for improving the reliability, energy efficiency, and sustainable operation of PV...
The training and evaluation of autonomous-driving perception systems rely on multimodal scene data. However, limited real-world long-tail scenarios have made generative world models an important complementary data source. Existing methods mainly...
The objective of this study was to compare subjective and objective visual quality after three types of corneal refractive surgery (LASEK, FS-LASIK, and TransPRK). 150 patients (300 eyes) who underwent corneal refractive surgery were divided into...
Multimodal biometric systems have gained increasing attention due to their ability to improve the reliability and security of user authentication. This study presents FI-VeT (Feature Fusion-based Iris-Finger Vein Vision Transformer), a lightweight...
Background Despite the widespread adoption of robotic systems for knee arthroplasty, the impact of prior robotic surgical experience on the transition to different robotic systems remains limited. Adaptation patterns across robotic platforms during...
To evaluate anterior corneal higher-order aberrations (HOAs) and inter-device agreement among Pentacam HR, ACE, and iTrace in primary congenital glaucoma (PCG). This cross-sectional study evaluated 37 eyes of 23 treated PCG patients at 3-, 4-, and...
Image retrieval has largely benefited from compact global descriptors and learned reranking, yet distinguishing hard candidates remains challenging when images share repeated structures, similar layouts, or partial visual overlap. In this work, we...
Purpose To present an automated method for computing the position of the RETeval skin sensor electrode and the iris color ratio (ICratio) from infrared eye images extracted from RETeval recording files. Methods Infrared eye images were extracted...
Reliable and precise machine perception demands fine spatial details to identify objects and high temporal resolution to track motion, which generates a data deluge that overwhelms computational resources. Overcoming this ‘data-wall’ challenge...
Ophthalmic & Physiological Optics(OPO), first published in 1925, is a leading international interdisciplinary journal that addresses basic and...
The International Journal of Computer Vision(IJCV) provides a forum for the dissemination of new research results in the rapidly growing field of...
Journal
The Visual Computerpublishes high-quality research on all aspects of computer graphics and visual computing, with a special focus on the rapidly...
