Conference Agenda
Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
|
Agenda Overview | |
|
Location: 713A 125 theatre |
| 8:30am - 5:00pm |
TuT1: 3D Reconstruction from Multi-View Satellite Imagery: From Classic to Modern Methods Location: 713A |
| 8:30am - 12:00pm |
TuT10: Open-source Scientific Software py4dgeo for Change Analysis in 3D/4D Point Clouds Location: 713A |
| 12:00pm - 1:15pm |
ICWG I/IV: Robotics for Mapping and Machine Intelligence Location: 713A A category-specific prompt strategy for semantic 3D indoor mapping using RGB-D camera 12:15pm - 12:30pm 3L-Planner: Lightweight LiDAR mapping and real-time local planning for ground robot autonomous navigation 12:30pm - 12:45pm CMCL-PR: Cross-Modal Camera-to-LiDAR Place Recognition with Cross-Attention Contrastive Learning |
| 1:30pm - 2:45pm |
WG II/4A: AI/ML for Geospatial Data Location: 713A Target Vessel Identification in Aerial Search Imagery via MLLM-Based Attribute Extraction and Geolocation Fusion 1:45pm - 2:00pm Zero-shot Vision-Language Reranking for Cross-View Geolocalization 2:00pm - 2:15pm From Pixels to Polygons: Evaluating Vision Foundation Models for High-Resolution Orthophoto Segmentation and Vectorization 2:15pm - 2:30pm Polarization-Aware Segmentation for Camouflaged Threat Detection from UAVs |
| 8:30am - 10:00am |
WG III/1A: Remote Sensing Data Processing and Understanding Location: 713A Cube Kernel: A Novel Approach to Enable Local Gradient Flow Across Channels in CNNs 8:45am - 9:00am Land Surface Dynamics Modeling and Prediction with dual Latent-Space Representations 9:00am - 9:15am Revealing Feature Contribution Mechanisms for Interpretable CNN-Transformer Remote Sensing Classification 9:15am - 9:30am EfficientViM-CD: An Efficient Remote Sensing Change Detection Network Based on Hidden State-Mixer 9:30am - 9:45am Local NMS: Enhancing Object Detection in Large-Scale Remote Sensing Images via iterative pipelined Postprocessing 9:45am - 10:00am ERD: Extended RAW-Diffusion Framework for De-rendering sRGB Images |
| 1:30pm - 3:00pm |
WG II/2A: Point Cloud Generation and Processing Location: 713A LGSSM: Local-to-global state space model for serialized point cloud semantic segmentation 1:45pm - 2:00pm Hierarchical Gaussian Partitioning for Semantic Segmentation of Airborne LiDAR Scenes 2:00pm - 2:15pm MCPF-Net: Multi-stage LiDAR-Image Collaborative Perception Fusion Network for Point Cloud Semantic Segmentation in Urban Scenes 2:15pm - 2:30pm Cross-Sensor Robustness and Spatial Generalization for 3D Railway Point Cloud Semantic Segmentation 2:30pm - 2:45pm Revisiting NeRF for Street Scene Point Cloud Semantic Segmentation in the Era of 3DGS 2:45pm - 3:00pm Extraction of Pole-like Road Objects from MMS Point Clouds Using Deep Learning and Geometric-Topological Feature Fusion |
| 3:30pm - 5:15pm |
ThS4A: Toward Smart Forests: Emerging Tools in Remote Sensing, Artificial Intelligence, and Field Robotics Location: 713A AI-Enabled Forest Inventory in TerraScan: integrating Georeferencing, Species Identification, and Volume Computation 3:45pm - 4:00pm Spatiotemporal Foundation Model for Aboveground Biomass Estimation: A case study in Mixedwood Plains Ecozone, Ontario, Canada 4:00pm - 4:15pm Improving Tree Species Detection for Operational Forestry: The Role of Dataset Design 4:15pm - 4:30pm Self-Supervised Leaf-Off Segmentation of Tree Functional Types and Buildings from Airborne NIRGB and LiDAR Data in Southern Ontario 4:30pm - 4:45pm Updating Forestry Road networks in Ontario Using Single Photon LiDAR and Deep Learning-enhanced algorithms 4:45pm - 5:00pm Attention-guided Multi-Scale Deep Learning Approach for Tree Health Detection Using Very High-Resolution Aerial Imagery 5:00pm - 5:15pm Fine-grained vegetation segmentation in complex urban park environments using a deeply supervised parallel SegFormer |
| 8:30am - 10:00am |
WG II/2B: Point Cloud Generation and Processing Location: 713A Multi-Source Fusion of Roof Skeletons, LiDAR and Street-View Imagery for Semi-Automated LoD-2 Building Modelling 8:45am - 9:00am BIM-to-Labelled Point Cloud : Automated Point Cloud Annotation from BIM Models using Bounding Boxes and Solid Geometry 9:00am - 9:15am Enhanced SegNet-based Building Extraction Framework via Image Segmentation and Point Cloud Fusion 9:15am - 9:30am Application Of Multi-Source Photogrammetric Data For Fast Building Inventory 9:30am - 9:45am Conjugate Feature-Guided Dense Stereo Matching for High-Precision Attribute-Enriched Urban Point Clouds 9:45am - 10:00am Efficient Extraction and Specification-Compliant Optimization of Railway Alignment Parameters from UAV LiDAR Point Clouds |
| 1:30pm - 3:00pm |
WG III/1B: Remote Sensing Data Processing and Understanding Location: 713A Multi-modal semantic segmentation for open vocabulary interactions with remote sensing images 1:45pm - 2:00pm Meta-Prompting with Open-Source Language Models for Zero-Shot Scene Classification in Remote Sensing 2:00pm - 2:15pm Knowledge graph enhanced for zero-shot semantic segmentation in remote sensing imagery 2:15pm - 2:30pm Segmentation-driven statistics-aware workflow for detailed scene description of UAV images using Mistral and LORA fused model 2:30pm - 2:45pm Evaluating the Adaptation Potential of SAM2 for Glacier Segmentation in Severe Weather 2:45pm - 3:00pm Reasoning-guided ego-path segmentation for autonomous trains using vision–language models |
| 3:30pm - 5:15pm |
ThS12: TLS-based Deformation Analysis Location: 713A Complementing and validating uncertainty of terrestrial laser scanning via interval analysis 3:45pm - 4:00pm Point-based, profile-based and 3D point cloud-based vibration monitoring of structures: comparisons based on a lab experiment 4:00pm - 4:15pm From tensor-product to truncated hierarchical B-splines: Enhancing spatial Resolution in space-continuous Deformation Analysis based on 3D point clouds 4:15pm - 4:30pm Towards a Framework for Benchmarking Dense 3D Displacement Estimation Approaches for Geomonitoring Using Long-Range TLS Data 4:30pm - 4:45pm Joint Stone Segmentation and Feature Driven Deformation Analysis at Water Dams 4:45pm - 5:00pm Reducing Non-rigidity in TLS Point Clouds Induced by Inhomogeneous Systematic Errors Using Free-form Surface Modeling 5:00pm - 5:15pm Calibration of Panoramic Terrestrial Laser Scanners using Planar Patches 5:15pm - 5:30pm Methodological framework for determining vertical angular variances of terrestrial laser scanners |
| 8:30am - 10:00am |
WG II/2C: Point Cloud Generation and Processing Location: 713A Differentiable deep consistency for point cloud registration 8:45am - 9:00am Cross-source Point Cloud Registration in the Bird’s-eye Domain: Aligning Street-level LiDAR with High-resolution Aerial Orthoimagery 9:00am - 9:15am Automated Alignment Enhancement of Backpack Image-LiDAR Data in a Forest Environment 9:15am - 9:30am A Marker-based Method for precise 3D Registration between CT-Data and photogrammetric Datasets 9:30am - 9:45am Advances in Historical Aerial Image Analysis: Boosting SfM Pipelines with Learned Models 9:45am - 10:00am Trinocular Multi-Object 3D Reconstruction in Camera-Simulating virtual Environments for Knee Arthroplasty |
| 1:30pm - 3:00pm |
WG III/1C: Remote Sensing Data Processing and Understanding Location: 713A Remote sensing image semantic segmentation sample generation using a decoupled latent diffusion framework 1:45pm - 2:00pm Bright-CC: A Novel Change Captioning Benchmark for Cross-Modal Remote Sensing Images 2:00pm - 2:15pm Remote Sensing Change/Damage Image Generator Based on Prior Foundation Model and Multimodal Reference Information 2:15pm - 2:30pm Edge Knowledge Distillation Guided Lightweight Change Detection Network 2:30pm - 2:45pm Leveraging Pretrained Priors for Weakly Supervised Semantic Segmentation of Remote Sensing Images 2:45pm - 3:00pm DeSEO: Physics-Aware Dataset Creation for High-Resolution Satellite Image Shadow Removal |
| 3:30pm - 5:15pm |
WG III/1D: Remote Sensing Data Processing and Understanding Location: 713A Spatio-temporal Modeling of Bridge Deformations from Sentinel-1 SAR Images Validated with Multiple In-situ Surveys 3:45pm - 4:00pm Large-Scale InSAR Deformation Monitoring Using Realistic Simulation-Based Training of a Temporal Convolutional Network: Application to the Phlegraean Fields, Italy 4:00pm - 4:15pm Geometry-conditioned Pix2Pix: leveraging explicit Conditioning on SAR projected local Incidence Angle for SAR-to-EO Translation Quality Improvement 4:15pm - 4:30pm Temporal-Spatial Tubelet Embedding for Cloud-Robust MSI Reconstruction using MSI-SAR Fusion: A Multi-Head Self-Attention Video Vision Transformer Approach 4:30pm - 4:45pm Evaluating Deep Matching Models for SAR-Optical Image Pairs using the SpaceNet9 Dataset 4:45pm - 5:00pm Detecting Marine Pollutants Using Sentinel-1 SAR and Sentinel-2 Optical Imagery 5:00pm - 5:15pm A coarse-to-fine cross-view localization framework with BEV-guided retrieval and fine-grained pose alignment |
| 8:30am - 10:00am |
WG III/1E: Remote Sensing Data Processing and Understanding Location: 713A Directional Total Least Square for FullWaveform Aerial LiDAR Smoothing 8:45am - 9:00am Improving Urban Point Cloud Classification Using Dynamic Local Context-Based Point Confidence 9:00am - 9:15am Refinenet: a confidence-aware deep online learning framework to refine real-world point cloud semantic segmentation 9:15am - 9:30am A Structured Query Language Approach for processing Smartphone-based LiDAR of Understory Vegetation 9:30am - 9:45am AI Indexing of Aerial LiDAR Point Cloud for Efficient Query 9:45am - 10:00am Intelligent Extraction Method for Geographic Information Feature Based on Human-Machine Collaboration |
| 1:30pm - 3:00pm |
WG II/2D: Point Cloud Generation and Processing Location: 713A An Approach for deriving Branch Kinematics of Deciduous Trees from hyper-temporal terrestrial Laser Scanner Data 1:45pm - 2:00pm In-Field 3D Wheat Head Instance Segmentation From TLS Point Clouds Using Deep Learning Without Manual Labels 2:00pm - 2:15pm Optimal Path Planning for Kinematic Laser Scanning 2:15pm - 2:30pm Non-Contact Modal Analysis of Wind Turbine Blades using Terrestrial Laser Scanner 2:30pm - 2:45pm Pixel-Accurate Registration of Photogrammetric Images and LiDAR in a Hybrid Airborne Oblique Imaging System 2:45pm - 3:00pm Integrating Airborne LiDAR and OpenStreetMap Features for Automated Hydrological Conditioning of Urban Digital Elevation Models |
| 3:30pm - 5:15pm |
SpS4A: Remote Sensing of Atmospheric Components for Climate Change and Air Quality: Bridging ISPRS and AERSS Location: 713A Satellite Remote Sensing and Numerical Simulation of the Impact of Biomass Burning on Black Carbon in East Asia 3:45pm - 4:00pm Validation of global land-ocean aerosol products retrieved from the DPC-2/GF-5(02) on-orbit measurements 4:00pm - 4:15pm Intra-urban aerosol heterogeneity in Hong Kong based on Lidar observations 4:15pm - 4:30pm Contrasting Meteorological Impacts of Dust Storms from the Gobi Desert versus the Taklimakan Desert over China 4:30pm - 4:45pm The Arctic Observing Mission (AOM): A high priority candidate mission for the Government of Canada 4:45pm - 5:00pm Global Point Source CO2 Emissions Monitoring Based on Hyperspectral Remote Sensing Imagery 5:00pm - 5:15pm Remote Sensing of CO, ozone and Their Correlation in Tropical Fire Regions |
| 8:30am - 10:00am |
WG II/2E: Point Cloud Generation and Processing Location: 713A Appearance-aware Scaling Diffusion Model for 3D Point Cloud Upsampling 8:45am - 9:00am Scan Outlier Ratio (ScOR): LiDAR Scanning and Survey-Aware Filtering of Detached Points in Terrestrial and Permanent Laser Scanning Point Clouds 9:00am - 9:15am LiDAR-Enhanced 3D Gaussian Splatting SLAM for Planetary Rover Exploration 9:15am - 9:30am Sensor Domain Adaptation for 3D Object Detection via LiDAR Super-Resolution 9:30am - 9:45am Ray Queries On Raw Point Clouds 9:45am - 10:00am Analysis of free large Area covering Elevation Models and improvement by ICESat-2 |
| 1:30pm - 3:00pm |
WG III/1F: Remote Sensing Data Processing and Understanding Location: 713A From Image to Perception: Scene-Graph-Driven Modeling of Human-Scale Urban Experience with Street-view Images 1:45pm - 2:00pm Real-Time Road Condition Detection and Mapping Using YOLOv11 and Built-In Car Dashcam 2:00pm - 2:15pm Towards Global Interpretability: Evaluating XAI Metrics in Building Footprint Extraction 2:15pm - 2:30pm MaskRoof: A deep Learning Framework and Benchmark Dataset for fine-grained urban Rooftop Utilization and potential Analysis 2:30pm - 2:45pm A comparison of CNN, Transformer, and open-vocabulary architectures for real-time photovoltaic defect detection using UAV thermal imagery. |
| 3:30pm - 5:15pm |
ThS18: Advances in Reality Capture, AI, and Digital Twin Technologies for Construction Engineering Location: 713A Image sequence based prediction of the temporal evolution of fresh concrete properties under realistic conditions 3:45pm - 4:00pm Single-image to model registration for semantic enrichment of indoor BIM 4:00pm - 4:15pm LSTNet: Local Shape Transformer Network for Road Marking Extraction 4:15pm - 4:30pm Automatic 3D Building Model Generation for Energy Digital Twins 4:30pm - 4:45pm From propagation to prediction: point-level uncertainty evaluation of MLS point clouds under limited ground truth 4:45pm - 5:00pm Automatic Scan-to-BIM: The Impact of Semantic Segmentation Accuracy on Opening Detection 5:00pm - 5:15pm Fast and accurate point surveying using the PIX4Dcatch mobile app |
| 8:30am - 10:00am |
ThS21: The Global-local Exchange Loop: Coupling Earth Observation and Citizen Sciences for LCLU Mapping Location: 713A OntoLULC-SOTA: An ontology based approach to make systematic reviews for LULC data 8:45am - 9:00am Manual Annotations meet Fine-Tuned Foundation Models: a Comparison on Tree Crown Segmentation Task 9:00am - 9:15am Evaluation of the IGN FLAIR-HUB Model Transferability Performance for Land Cover Mapping in Iasi, Romania 9:15am - 9:30am Towards efficient Giant Tree Inventories: Deep Learning with crowdsourced Training Data 9:30am - 9:45am The Global-Local loop: what is missing in bridging the gap between geospatial data from numerous communities ? |
| 10:30am - 12:00pm |
WG III/1G: Remote Sensing Data Processing and Understanding Location: 713A YOLOv8m-CCFM-GSConv: Research on Lightweight Marine Oil Spill Target Detection Based on Improved YOLOv8m Model 10:45am - 11:00am Detecting moving vehicles on Sentinel-2 imagery using semi-automatic labeling from S2A/S2C tandem phase 11:00am - 11:15am LAD-Enhancer: A Lightweight All in One Aerial Detection Enhancer Under Adverse Weather 11:15am - 11:30am A Collaborative Detection Method of Small Unmanned Aerial Vehicle Target via Multi-modal Feature Fusion in Complex Background 11:30am - 11:45am Infrared-Visible Image Fusion Method Based on Differential Feature Enhancement and Cross-Modal Attention |
