Fumio Okura / 大倉 史生
  • Bio
  • Publications
  • Talks
  • News
  • Experience
  • Projects
  • Publications
    • Don't mask out the background! Natural-light photometric stereo via illumination reconstruction
    • NRGS: Neural regularization for robust 3D semantic Gaussian splatting
    • High-fidelity multi-view normal integration with scale-encoded neural surface representation
    • BioVITA: Biological dataset, model, and benchmark for visual-textual-acoustic alignment
    • Empirical hyperelastic modeling of textile fabrics for measuring nonlinear bending rigidity
    • PlantPose: Universal plant skeleton estimation via tree-constrained graph generation
    • Unsupervised 3D human pose estimation via conditional multi-view ancestral sampling
    • DP-SfM: Dual-pixel structure-from-motion without scale ambiguity
    • AnimalCLAP: Taxonomy-aware language-audio pretraining for species recognition and trait inference
    • Gaussian mesh renderer for lightweight differentiable rendering
    • Instance-wise distribution control of text-to-image diffusion models
    • Talking with Actionbits---A part-enhanced VLM for action and interaction recognition in animals
    • Interaction-via-Actions: Cattle interaction detection with joint learning of action-interaction latent space
    • Zero-shot hierarchical plant segmentation via foundation segmentation models and text-to-image attention
    • NeuraLeaf: Neural parametric leaf models with shape and deformation disentanglement
    • Spectral sensitivity estimation with an uncalibrated diffraction grating
    • LeafGen: Structure-aware leaf image generation for annotation-free leaf instance segmentation
    • Masks-to-Skeleton: Multi-view mask-based tree skeleton extraction with 3D Gaussian splatting
    • A multi-modality fusion model based on dual-task measurement for the automatic detection of early-stage cognitive impairment
    • HoGS: Unified near and far object reconstruction via homogeneous gaussian splatting
    • The clinical significance of an AI-based assumption model for neurocognitive diseases using a novel dual-task system
    • TreeFormer: Single-view plant skeleton estimation via tree-constrained graph generation
    • Predicting future cognitive decline from long-term observations of dual-task performance data
    • Predicting future cognitive decline from long-term observations of dual-task performance data
    • Resolving scale ambiguity in multi-view 3D reconstruction using dual-pixel sensors
    • Synthesizing time-varying BRDFs via latent space
    • Path-space differentiable rendering of implicit surfaces
    • MVCPS-NeuS: Multi-view constrained photometric stereo for neural surface reconstruction
    • NeRSP: Neural 3D reconstruction for reflective objects with sparse polarized images
    • PPGCN: Phase-aligned periodic graph convolutional network for dual-task-based cognitive impairment detection
    • Predicting molecular interactions by graph convolutional neural networks with global features
    • Computer vision in smart city application: A mapping review
    • M3R-CNN: On effective multi-modal fusion of RGB and depth cues for instance segmentation in bin-picking
    • Learn to synthesize photorealistic dual-pixel images from RGBD frames
    • Near-light photometric stereo with symmetric lights
    • Two-stream graph convolutional networks with task-specific loss for dual-task gait analysis
    • Multi-View azimuth stereo via tangent space consistency
    • Accuracy evaluation and prediction of single-image camera calibration
    • Shuffled linear regression with outliers in both covariates and responses
    • Bilateral normal integration
    • Discrete search photometric stereo for fast and accurate shape estimation
    • Multispectral photometric stereo for spatially-varying spectral reflectances
    • Shape and albedo recovery by your phone using stereoscopic flash and no-flash photography
    • Exploring the mechanical and morphological rationality of tree branch structure based on 3D point cloud analysis and the finite element method
    • 3D modeling and reconstruction of plants and trees: A cross-cutting review across computer graphics, vision, and plant phenotyping
    • Close-contact detection using a single camera for sports considering occlusion
    • Detecting lower MMSE scores in older adults using cross-trial features from a dual-task with gait and arithmetic
    • Shape-coded ArUco: Fiducial marker for bridging 2D and 3D modalities
    • Symmetric-light photometric stereo
    • Human localization using a single camera towards social distance monitoring during sports
    • Early detection of low cognitive scores from dual-task performance data using a spatio-temporal graph convolutional neural network
    • A closer look at rotation-invariant deep point cloud analysis
    • Generalized shuffled linear regression
    • Multispectral photometric stereo for spatially-varying spectral reflectances: A well posed problem?
    • Normal integration via inverse plane fitting with minimum point-to-plane distance
    • Descriptor-free multi-view region matching for instance-wise 3D reconstruction
    • Training instance segmentation neural network with synthetic datasets for crop seed phenotyping
    • Statistical analysis of dual-task gait characteristics for cognitive score estimation
    • A unified framework for image-based plant phenotyping under controlled growth environment: From image acquisition to phenotype interpretation
    • Comparison of 3D reconstructions from UAV RGB images and terrestrial LiDAR imaging for measuring architectural traits of peach trees
    • Toward 3D plant modeling of every leaf and branchlet
    • RGB-D video-based individual identification of dairy cows using gait and texture analyses
    • Multi-view instance matching for plant leaf modeling
    • How convolutional neural networks diagnose plant disease
    • Early detection of lower MMSE scores in elderly based on dual-task gait
    • Estimating 3D human shape under clothing from a single RGB image
    • Seeing behind leaves: Multi-view reconstruction of three-dimensional branch structure
    • Growth assessment of school-age children from dual-task observation
    • Directional characteristics evaluation of silhouette-based gait recognition
    • Probabilistic plant modeling via multi-view image-to-image translation
    • Automatic collection of dual-task human behavior for analysis of cognitive function
    • Age estimation from dual-task behavior for comprehensive growth assessment of children
    • Realtime novel view synthesis with eigen-texture regression
    • Motion parallax representation for indirect augmented reality
    • Gait Collector: An automatic gait data collection system in conjunction with an experience-based long-run exhibition
    • Addressing temporal inconsistency in indirect augmented reality
    • Unifying color and texture transfer for predictive appearance manipulation
    • Unifying color and texture transfer for predictive appearance manipulation
    • Mixed-reality world exploration using image-based rendering
    • Indirect augmented reality considering real-world illumination change
    • Aerial full spherical HDR imaging and display
    • Background estimation for a single omnidirectional image sequence captured with a moving camera
    • Free-viewpoint mobile robot teleoperation interface using view-dependent geometry and texture
    • Mobile AR using pre-captured omnidirectional images
    • Teleoperation of mobile robots by generating augmented free-viewpoint images
    • Full spherical high dynamic range imaging from the sky
    • Interactive exploration of augmented aerial scenes with free-viewpoint image generation from pre-rendered images
    • Fly-through Heijo Palace Site: Historical tourism system using augmented telepresence
    • Spacetime freeview generation using image-based rendering, relighting, and augmented telepresence
    • Tone mapping for HDR images with dimidiate luminance and spatial distributions of bright and dark regions
    • Fly-through Heijo Palace Site: augmented telepresence using aerial omnidirectional videos
    • Augmented telepresence using autopilot airship and omni-directional camera
  • Blog
    • 🎉 A paper was accepted to ECCV 2026!
    • 🎉 A paper was accepted to CVPR 2026!
    • 🎉 Two papers were accepted to WACV 2026!
    • 🎉 Two papers were accepted to ICCV 2025!
    • 🎉 A paper was accepted to CVPR 2025!
    • 🎉 A paper was accepted to WACV 2025!
    • 🎉 Two papers were accepted to ECCV 2024!
    • 🎉 A paper was accepted to SIGGRAPH 2024!
  • Recent & Upcoming Talks
    • NeuraLeaf: 葉のニューラルパラメトリックモデリング @ 日本生物環境工学会
    • Panel Discussion: The Science of Phenomics @ MLCAS2025
    • HoGS: Unified near and far object reconstruction via homogeneous gaussian splatting @ MIRU2025
    • 画像からの植物「構造」の復元 @ 第31回食料生産技術研究会
    • 緻密な植物デジタルツイン生成に向けた植物構造復元 @ 日本生物環境工学会
    • Agricultural development using digital technologies and best practices from Japan @ Thai
    • 3D structural reconstruction of plants: A perspective from computer vision study @ MLCAS2022
    • 画像からの枝葉「構造」の復元 @ 日本植物バイオテクノロジー学会
    • 農業は情報の「応用」か？農業分野から新しい情報技術を生み出す @ 農業情報学会
    • 画像からの枝葉構造の定量化 @ 日本生態学会
    • 多視点画像からの植物の三次元構造復元に向けて @ 農業情報学会
    • コンピュータビジョンにおける植物 @ 日本生態学会
    • AIと画像解析 @ 日本育種学会
    • 人工知能x植物科学：そもそもAIってなんだ @ 日本植物学会
    • Automatically acquiring walking-related behavior of 100,000 people @ International Workshop on Human Behavior Analysis in the Real World
  • Projects
    • Plant Twin
    • 3D Reconstruction
  • Projects
  • Experience
  • Teaching
    • Learn JavaScript
    • Learn Python

🎉 A paper was accepted to ECCV 2026!

Jul 15, 2026·
Fumio Okura
Fumio Okura
· 1 min read

🎉 Congratulations, Hashida-san! We will present the following paper in ECCV 2026 Malmo!

  • Taiga Hashida, Hiroaki Santo, Fumio Okura. Don’t mask out the background! Natural-light photometric stereo via illumination reconstruction. In ECCV 2026.
Last updated on Jul 15, 2026
ECCV 2026 Computer Vision
Fumio Okura
Authors
Fumio Okura
Associate Professor
🎉 A paper was accepted to CVPR 2026! Feb 24, 2026 →

© 2026 Fumio Okura.

Published with Hugo Blox Builder — the free, open source website builder that empowers creators.