JA EN
← Classic CV↗

Ph.D. Student — Keio University

Wataru Ikeda池田 航

Seeing human motion in 3D through event cameras, sensors that perceive nothing but change.

Move your cursor: only pixels that change light up. Click to flash.

Scroll

Latest

  1. IEEE Access · JournalEventEgoHands++: Event-based Egocentric 3D Hand Mesh Reconstruction with Real Dataset
  2. ACCV 2026 · InternationalOptical Music Recognition for Handwritten Kateishiki-fu Scores using Symbol Detection and Melody Completion
  3. MIRU2026 · AwardMIRU Student Encouragement Award

(01)

AboutProfile

Portrait of Wataru Ikeda
池田 航 Wataru Ikeda

Second-year doctoral student at Keio University Graduate School, working in Computer Vision and Deep Learning. I use event cameras to estimate egocentric 3D human pose and reconstruct hand shape.

Affiliation
Ph.D. program (2nd year), Keio Universityas of Oct 2026
Field
Computer Vision, Deep Learning
Email
wataru-ikeda[at]keio[dot]jp
  • Event Camera
  • Egocentric Vision
  • 3D Human Pose
  • Hand Mesh
  • IMU Fusion
  • Sports Sensing
  • 04Awards
  • 04International papers
  • 10Domestic papers
  • 02Joint research

(02)

AwardsRecognition

  1. MIRU Student Encouragement Award

    “Egocentric 3D Human Pose Estimation using Event-based Camera and IMUs”

    Wataru Ikeda, Ryosuke Hori, Mariko Isogawa, Vladislav Golyanik

    The 29th Symposium on Sensing via Image Information MIRU2026

    33.5%

    Given to the student first authors of oral papers in the top 33.5% (116 of 346 submissions), selected through peer review.

  2. MIRU Student Encouragement Award

    “Event-based Egocentric Human Pose Estimation in Dynamic Environment”

    Wataru Ikeda, Mariko Isogawa

    The 28th Symposium on Sensing via Image Information MIRU2025 · Awards | MIRU2025 ↗

    34.5%

    Given to the student first authors of oral papers in the top 34.5% (88 of 255 submissions), selected through peer review.

  3. MIRU Interactive Presentation Award

    “D-EventEgo: Egocentric 3D Human Pose Estimation Using Event Camera in Dynamic Environment”

    Wataru Ikeda, Mariko Isogawa

    The 27th Symposium on Sensing via Image Information MIRU2024 · Awards | MIRU2024 ↗

    4.3%

    22 of 517 poster presentations received the award.

  4. MIRU Demo Presentation Award

    “EventPointMesh in Realtime: Real-time 3D Human Pose and Shape Estimation Using Only Event Information”

    Wataru Ikeda*, Hiroyuki Deguchi*, Ryosuke Hori, Mariko Isogawa, Hideo Saito

    The 26th Symposium on Sensing via Image Information MIRU2023 · Awards | MIRU2023 ↗

    1/10

    1 of 10 demo presentations received the award.

* Equal contribution

(03)

PublicationsPapers & Talks

  1. 2026

    EventEgoHands++: Event-based Egocentric 3D Hand Mesh Reconstruction with Real Dataset

    Ryosei Hara, Wataru Ikeda, Masashi Hatano, Mariko Isogawa

    IEEE Access, 2026 (Early Access)

    • Journal
    • Impact Factor 4.2

  2. 2026

    Optical Music Recognition for Handwritten Kateishiki-fu Scores using Symbol Detection and Melody Completion

    Emika Kurita, Wataru Ikeda, Shuhei Yamada, Haruyoshi Suzuki, Maki Sugimoto, Mariko Isogawa

    ACCV 2026 · Osaka, Japan

    • International
    • h5-index 52

  3. 2026

    Event-Based Batting Impact Estimation

    Ryotaro Ishida, Wataru Ikeda, Ryosei Hara, Akemi Kobayashi, Toshitaka Kimura, Mariko Isogawa

    ICIP 2026 · Tampere, Finland

    • International
    • h5-index 55
    • Acceptance 38.6%

  4. 2025

    Event-based Egocentric Human Pose Estimation in Dynamic Environment↗

    Wataru Ikeda, Masashi Hatano, Ryosei Hara, Mariko Isogawa

    ICIP 2025

  5. 2025

    EventEgoHands: Event-based Egocentric 3D Hand Mesh Reconstruction↗

    Ryosei Hara, Wataru Ikeda, Masashi Hatano, Mariko Isogawa

    ICIP 2025

  6. 2026

    Optical Music Recognition of Handwritten Kateishiki-fu Scores Using Symbol Recognition and Melody Correction

    Emika Kurita, Wataru Ikeda, Shuhei Yamada, Haruyoshi Suzuki, Maki Sugimoto, Mariko Isogawa

    Hokkaido Information University, Ebetsu, Hokkaido, Japan · Sep 1, 2026

    • Oral

  7. 2026

    Egocentric 3D Human Pose Estimation using Event-based Camera and IMUs

    Wataru Ikeda, Ryosuke Hori, Mariko Isogawa, Vladislav Golyanik

    The 29th Symposium on Sensing via Image Information MIRU2026 · Dejima Messe Nagasaki, Nagasaki, Japan · Aug 5, 2026

    • Peer-reviewed oral
    • Top 33.5%
    • Student Encouragement Award

  8. 2026

    3D Human Body and Hand Shape Reconstruction Using Event Cameras

    Wataru Ikeda, Ryosei Hara, Hiroyuki Deguchi, Masashi Hatano, Ryosuke Hori, Hideo Saito, Mariko Isogawa

    The 40th Annual Conference of the Japanese Society for Artificial Intelligence · G Messe Gunma, Gunma, Japan · Jun 11, 2026

    • Organized session oral

  9. 2026

    Robust Impact Timing Estimation in Sports under Low-Light and Occluded Conditions

    Ryotaro Ishida, Wataru Ikeda, Ryosei Hara, Akemi Kobayashi, Satoki Kimura, Mariko Isogawa

    The 4th Sports Informatics Research Conference · UEC Tokyo, Tokyo, Japan · Mar 4, 2026

    • Oral

  10. 2025

    Event-based Egocentric Human Pose Estimation in Dynamic Environment

    Wataru Ikeda, Masashi Hatano, Ryosei Hara, Mariko Isogawa

    The 28th Symposium on Sensing via Image Information MIRU2025 · KICC, Kyoto, Japan · Jul 31, 2025

    • Peer-reviewed oral
    • Top 34.5%
    • Student Award

  11. 2025

    EventEgoHands: Event-based Egocentric 3D Hand Mesh Reconstruction

    Ryosei Hara, Wataru Ikeda, Masashi Hatano, Mariko Isogawa

    The 28th Symposium on Sensing via Image Information MIRU2025 · KICC, Kyoto, Japan · Jul 31, 2025

    • Peer-reviewed oral
    • Top 34.5%

  12. 2025

    Impact Timing Estimation in Batting Using an Event Camera

    Ryotaro Ishida, Wataru Ikeda, Ryosei Hara, Akemi Kobayashi, Satoki Kimura, Mariko Isogawa

    Sports Informatics and Technology (SIT) · Tokyo Big Sight, Tokyo, Japan · Aug 1, 2025

    • Oral

  13. 2024

    D-EventEgo: Egocentric 3D Human Pose Estimation Using Event Camera in Dynamic Environment

    Wataru Ikeda, Mariko Isogawa

    The 27th Symposium on Sensing via Image Information MIRU2024 · Kumamoto Castle Hall, Kumamoto, Japan · Aug 7, 2024

    • Interactive Presentation Award

  14. 2024

    3D Human Hand Shape Estimation Using Event Camera under Dynamic Background

    Ryosei Hara, Wataru Ikeda, Mariko Isogawa

    The 27th Symposium on Sensing via Image Information MIRU2024 · Kumamoto Castle Hall, Kumamoto, Japan · Aug 7-8, 2024

    • Poster

  15. 2023

    EventPointMesh in Realtime: Real-time 3D Human Pose and Shape Estimation Using Only Event Information

    Wataru Ikeda*, Hiroyuki Deguchi*, Ryosuke Hori, Mariko Isogawa, Hideo Saito

    The 26th Symposium on Sensing via Image Information MIRU2023 · Act City Hamamatsu, Shizuoka, Japan · Jul 26-27, 2023

    • Demo Award

* Equal contribution

(04)

ExperienceInternships & Freelance

2024.08 — 2024.09 NEC Corporation Research Internship
  • Improving spatiotemporal prediction accuracy in world models

Reference — “World Model” Further Evolves: Robot AI Technology that Adapts to Environments and Makes Precise Movements ↗

2024.07 — 2024.08 CyberAgent, Inc. AI Business Division, Extreme Prediction AI Team — Data Science
  • Improving inference accuracy for advertising effectiveness using video input
  • Deep Learning
  • Python
  • GCP

Reference — Revolutionizing AI-based Creative Production: Launch of “Extreme Prediction AI” with Performance-based Pricing ↗
AI/Data Technology Map (pdf), Extreme Prediction AI ↗

2022.03 — 2024.05 Lightblue Co., Ltd. Engineer — Vision, Cloud, LLM
  • Image processing (Object detection, OCR, Deep Learning, PyTorch)
  • Video processing (Gaze estimation, Deep Learning, PyTorch)
  • AWS (Lambda, EC2, SageMaker, ECS, DynamoDB etc.)
  • Edge AI computing (NVIDIA Jetson etc.)
  • LLM (OpenAI API, Azure OpenAI Service etc.)
  • Web development (Vue, React, TypeScript)
  • Teams app development (FastAPI, Python)
2023.06 — 2025.10 Freelance AI implementation and improvement in IT, marketing, pharmaceutical, and real estate
  • Demand forecasting (Regression models, Time series models, LightGBM, Ensemble methods)
  • Data analysis (Principal component analysis, Cosine similarity, Co-occurrence matrix, Granger causality test, Basic statistical analysis)
  • Web application and API development for prediction models and data analysis (Streamlit, Google Cloud Run)
  • Azure Chat web development (Next.js, TypeScript)
  • Image processing (Deep Learning, Instance Segmentation)
  • Image generation (Deep Learning, Diffusion Models, Azure Machine Learning)
  • Zenn book writing collaboration — Quick Start Chatbot【ChatGPT / Python】↗ (500+ likes), Quick Start Prompt Engineering【LangChain / ChatGPT】↗ (1000+ likes)

(05)

Grants & MoreFunding, Activities, Qualifications

Grants Non-repayable

  • JST BOOST / Keio-BOOST2025.09 — 2027.03¥3,600,000 / yr+ ¥300,000 research fund
  • Keio University Fujiwara Scholarship Fund2025.09 — 2026.03¥500,000
  • Keio University “Susume” Scholarship2026.06¥300,000

Loans Repayable

  • JASSO Type I Scholarship2024.04 — 2025.09¥88,000 / mo

A

Research ActivitiesCollaboration & Talks

  • Joint research on scene graphs using visual information with Prof. Satoshi Kurihara (President of JSAI), Keio University
  • Joint research on pose estimation with Dr. V. Golyanik (PI, Max Planck Institute, Germany)
  • Presented a survey on geometry research for event cameras and agriculture at MIRU2025 Young Researchers Program

B

OthersMilestones

  • Early completion of Master's program at Keio University Graduate School

C

QualificationsCertifications

  • Fundamental Information Technology Engineer Examination
  • Certified Expert of Image Processing Engineers