Computer Vision

arXiv:2605.01483v1 Announce Type: cross Abstract: A hierarchical cross-modal fusion model is proposed for vision-language question answering (VLQA) in industria

arXiv:2605.01490v1 Announce Type: cross Abstract: Pansharpening aims to generate high-resolution multispectral (HRMS) images by fusing low-resolution multispect

arXiv:2605.01668v1 Announce Type: cross Abstract: Dense temporal annotation of procedural activity videos is vital for action understanding and embodied intelli

arXiv:2605.01876v1 Announce Type: cross Abstract: Multimodal resources for non-periodic court sports with laboratory-grade sensing remain scarce: few publicly p

arXiv:2605.02086v1 Announce Type: cross Abstract: 3D Gaussian splatting (3DGS) is a state-of-the-art representation for real-time photorealistic novel-view synt

arXiv:2605.02137v1 Announce Type: cross Abstract: Accurate flood water mapping is critical for disaster management, yet current methods struggle to fully exploi

arXiv:2605.02153v1 Announce Type: cross Abstract: Synthetic Aperture Radar (SAR) imagery is widely used for flood monitoring due to its all-weather and day-nigh

arXiv:2605.02184v1 Announce Type: cross Abstract: Pansharpening aims to generate high-resolution multispectral (HRMS) images by fusing low-resolution multispect

arXiv:2605.02537v1 Announce Type: cross Abstract: Autonomous 3D indoor scene synthesis breaks down in non-convex rooms with tightly coupled spatial constraints.

arXiv:2605.02638v1 Announce Type: cross Abstract: Cross-view Referring Multi-Object Tracking (CRMOT) aims to track multiple objects specified by natural languag

arXiv:2303.10971v2 Announce Type: replace-cross Abstract: The matching of 3D shapes has been extensively studied for shapes represented as surface meshes, as we

arXiv:2410.18717v2 Announce Type: replace-cross Abstract: Recent advancements in artificial intelligence hold ample potential for monitoring applications using

arXiv:2504.17401v2 Announce Type: replace-cross Abstract: Stereo disparity estimation is crucial for obtaining depth information in robot-assisted minimally inv

arXiv:2507.21114v3 Announce Type: replace-cross Abstract: Digitization projects in humanities often generate vast quantities of page images from historical docu

arXiv:2511.19474v5 Announce Type: replace-cross Abstract: Automatically detecting abnormal events in videos is crucial for modern autonomous systems, yet existi

arXiv:2601.17756v3 Announce Type: replace-cross Abstract: Existing Subject-to-Video Generation (S2V) methods have achieved high-fidelity and subject-consistent

arXiv:2601.23286v2 Announce Type: replace-cross Abstract: While recent video diffusion models (VDMs) produce visually impressive results, they fundamentally str

Computer vision holds incredible potential when it comes to observing animals in their natural environments. This article provides a… Continue reading on Object

YOLOv8 nano + CoreML + Apple Neural Engine — building an offline-first livestock counter and what I learned about confidence thresholds, coordinate flips, and e

arXiv:2604.26991v1 Announce Type: cross Abstract: Recent advances in data-centric medical AI have produced highly accurate diagnostic systems, but the emphasis

arXiv:2604.27106v1 Announce Type: cross Abstract: Accurately reconstructing complex full multi-object scenes from sparse observations remains a core challenge i

arXiv:2604.27617v1 Announce Type: cross Abstract: With the widespread application of Unmanned Aerial Vehicles (UAVs) in bridge structural health monitoring, dee

arXiv:2604.27928v1 Announce Type: cross Abstract: Tunnel inspection requires outputs that can support defect localization, measurement, severity grading, and en

arXiv:2604.27975v1 Announce Type: cross Abstract: Traditional Shot Boundary Detection (SBD) inherently struggles with complex transitions by formulating the tas

arXiv:2604.28169v1 Announce Type: cross Abstract: Modern video diffusion models excel at appearance synthesis but still struggle with physical consistency: obje

arXiv:2507.12414v2 Announce Type: replace-cross Abstract: Training of autonomous driving systems requires extensive datasets with precise annotations to attain

arXiv:2509.25845v3 Announce Type: replace-cross Abstract: Recent advancements in diffusion and flow-matching models have demonstrated remarkable capabilities in

Growing up playing playing baseball, I did what any amateur hitter would do: record my swing, analyze my mechanics, and attempt to… Continue reading on Medium »

No sliders. No presets. Just math that listens to your image. If you've ever used CLAHE (Contrast...

Most OCR tools are built for clean text. Receipts. Documents. Screenshots. Menus. Maybe a street sign if the lighting is kind. Manga is none of those things. A

Open-Vocabulary Foundation Models, Deep Learning, and Classical Geometry into a Unified Object Detection Library Continue reading on Stackademic »

Every day, banking ops teams manually review thousands of documents - loan applications, KYC forms,...

Every day, banking ops teams manually review thousands of documents - loan applications, KYC forms,...

I built facecalculators.com a free, privacy-first face shape detector that runs entirely in the...

A ground truth guide to how cameras distort reality and why calibration is critical for accurate computer vision systems. Continue reading on Medium »

A ground truth guide to how cameras distort reality and why calibration is critical for accurate computer vision systems. Continue reading on Medium »

arXiv:2604.26084v1 Announce Type: cross Abstract: Accurate fruit maturity identification is essential for determining harvest timing, as incorrect assessment di

arXiv:2604.26221v1 Announce Type: cross Abstract: Open-vocabulary semantic segmentation (OVSS) in remote sensing images is a promising task that employs textual

arXiv:2604.26244v1 Announce Type: cross Abstract: We study generative super-resolution (SR) in real-world scenarios where content and degradations vary across d

arXiv:2604.26435v1 Announce Type: cross Abstract: The rapid advancement of object detection architectures has positioned single stage detectors as the dominant

arXiv:2604.26633v1 Announce Type: cross Abstract: The bottleneck in learning-based industrial defect detection is often limited not by model capacity, but by th

arXiv:2604.26774v1 Announce Type: cross Abstract: Open-vocabulary change detection aims to identify semantic changes in bi-temporal remote sensing images withou

arXiv:2604.26806v1 Announce Type: cross Abstract: Transformer-based architectures have established a dominant paradigm in global semantic perception; however, t

arXiv:2508.09547v2 Announce Type: replace-cross Abstract: We introduce Goal-Conditioned Visual Navigation Instruction Generation (GoViG), a new task that aims t

arXiv:2512.08982v2 Announce Type: replace-cross Abstract: Retinex-based low-light image enhancement benefits from separating reflectance and illumination, yet r

How do convolutional Neural Networks in Machine Learning and a Historical Heritage founded in 1574 connect? Continue reading on Medium »

Join us on May 1 for day two of the Best of WACV 2026 series of virtual events. Register for the...

Analyzing the geometric failure points of synthetic media For developers working in computer vision...