Results 81 to 90 of about 10,859 (240)
MRPose: Multi‐Robot Relative 6D Pose Estimation for Unseen Objects From RGB Images
ABSTRACT Relative pose estimation under the query–reference paradigm has emerged as a practical alternative for estimating the 6D pose of unseen objects without relying on CAD models or extensive annotations. However, many existing approaches remain difficult to deploy in practice, as their reliance on geometric matching makes robustness highly ...
Jiale Ren +4 more
wiley +1 more source
Fast Feature-Based Template Matching, Based on Efficient Keypoint Extraction
In order to improve the performance of feature-based template matching techniques, several research papers have been published. Real-time applications require the computational complexity of keypoint matching algorithms to be as low as possible.
Mohammad Soltanian, Ali Tavakol
core +1 more source
ABSTRACT Infrared and visible image registration is a key task in automated distribution network inspection. Previous approaches typically apply a single homography to align infrared and visible images leading to local misalignment. To address this, this study proposes a novel region‐based registration algorithm specifically designed for wide‐depth ...
Yingchen Zhang, Bo Wang, Fei Tang
wiley +1 more source
Keypoint label template of the cow skeleton.
Keypoint label template of the cow skeleton.
Zhi Weng (8451915) +5 more
core +1 more source
ABSTRACT Unmanned Aerial Vehicle (UAV) inspection of overhead transmission line composite insulators allows for the rapid acquisition of large amounts of infrared images of insulators in service, which can be used to analyse their operational status. However, efficiently and accurately analysing these images remains a challenge.
Zhenan Zhou +7 more
wiley +1 more source
KINet: Keypoint Interaction Networks for Unsupervised Forward Modeling
Object-centric representation is an essential abstraction for forward prediction. Most existing forward models learn this representation through extensive supervision (e.g., object class and bounding box) although such ground-truth information is not ...
Rezazadeh, Alireza, Choi, Changhyun
core
ABSTRACT With the development of electric power artificial intelligence (AI) technology, many complicated challenges emerged during application. Traditional AI algorithms work well in specialised tasks such as detection and classification. However, they are unlikely to solve general problems.
Yuanpeng Tan +4 more
wiley +1 more source
Doodle Your Keypoints: Sketch-Based Few-Shot Keypoint Detection
Keypoint detection, integral to modern machine perception, faces challenges in few-shot learning, particularly when source data from the same distribution as the query is unavailable. This gap is addressed by leveraging sketches, a popular form of human expression, providing a source-free alternative.
Subhajit Maity +5 more
openaire +2 more sources
A validated 3D point‐cloud analysis workflow for the creation of seagrass canopy height models
Abstract Seagrasses support highly productive ecosystems in both temperate and tropical waters and provide essential ecosystem services. Despite their importance, the vertical structure of seagrasses (i.e. the canopy) remains poorly characterised at the microscale, even though it is a strong predictor of ecological processes that influence both ...
Daniele Ventura
wiley +1 more source
Abstract The biodiversity crisis demands new approaches to describing and characterising species that can scale beyond the capacity of professional taxonomists alone. Hundreds of millions of specimen images are now openly available, yet almost none carry structured information about the morphology they depict. We present the Descriptron‐GBIF Annotator,
Alex R. Van Dam, Francisco Hita Garcia
wiley +1 more source

