Gui-Song Xia

42

Papers

48

Total Citations

Papers (42)

FreePoint: Unsupervised Point Cloud Instance Segmentation

NEAT: Distilling 3D Wireframes from Neural Attraction Fields

Anchor-based Robust Finetuning of Vision-Language Models

UNIC-Adapter: Unified Image-instruction Adapter with Multi-modal Transformer for Image Generation

Holistic Large-Scale Scene Reconstruction via Mixed Gaussian Splatting

NeurIPS 2025arXiv

Fuse Before Transfer: Knowledge Fusion for Heterogeneous Distillation

Patched Line Segment Learning for Vector Road Mapping

Unleashing Unlabeled Data: A Paradigm for Cross-View Geo-Localization

3D Building Reconstruction from Monocular Remote Sensing Images with Multi-level Supervisions

DMTG: One-Shot Differentiable Multi-Task Grouping

DOTA: A Large-Scale Dataset for Object Detection in Aerial Images

Learning Attraction Field Representation for Robust Line Segment Detection

Learning to Calibrate Straight Lines for Fisheye Image Rectification

Learning RoI Transformer for Oriented Object Detection in Aerial Images

Zero-Assignment Constraint for Graph Matching With Outliers

Holistically-Attracted Wireframe Parsing

FGN: Fully Guided Network for Few-Shot Instance Segmentation

ReDet: A Rotation-Equivariant Detector for Aerial Object Detection

Deep Graph Matching Under Quadratic Constraint

Event-Based Synthetic Aperture Imaging With a Hybrid Network

Expanding Low-Density Latent Regions for Open-Set Object Detection

Learning Local-Global Contextual Adaptation for Multi-Person Pose Estimation

Revisiting Document Image Dewarping by Grid Regularization

Decoupling Zero-Shot Semantic Segmentation

HGFormer: Hierarchical Grouping Transformer for Domain Generalized Semantic Segmentation

OmniCity: Omnipotent City Understanding With Multi-Level and Multi-View Images

Dynamic Coarse-To-Fine Learning for Oriented Tiny Object Detection

PlaneTR: Structure-Guided Transformers for 3D Plane Recovery

Motion Deblurring With Real Events

3D Building Reconstruction From Monocular Remote Sensing Images

Parsing Table Structures in the Wild

Sat2Density: Faithful Density Learning from Satellite-Ground Image Pairs

Generalizing Event-Based Motion Deblurring in Real-World Scenarios

Event Enhanced High-Quality Image Recovery

RFLA: Gaussian Receptive Field Based Label Assignment for Tiny Object Detection

Level-S$^2$fM: Structure From Motion on Neural Level Set of Implicit Surfaces

Exploring Scene Affinity for Semi-Supervised LiDAR Semantic Segmentation

Beyond Pixel Uncertainty: Bounding the OoD Objects in Road Scenes

AdaDCP: Learning an Adapter with Discrete Cosine Prior for Clear-to-Adverse Domain Generalization

Learning Fine-grained Domain Generalization via Hyperbolic State Space Hallucination

VHM: Versatile and Honest Vision Language Model for Remote Sensing Image Analysis

ConDaFormer: Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding