Pattern Recognition Laboratory

広島大学
パターン認識研究室

Pattern Recognition Laboratory, Hiroshima University

Members / Research / Contact / Aizawa Profile

Overview

広島大学 パターン認識研究室のWebサイトです．栗田多喜夫先生が運営していたパターン認識研究室 （栗田・宮尾・相澤研究室）を相澤宏旭が引き継ぎ，2024年から新たにスタートした研究室です．研究室では， 主に画像認識・画像生成のための機械学習・深層学習の基礎技術について研究しています．

This is the website of the Pattern Recognition Laboratory at Hiroshima University. Our group mainly studies foundational machine learning and deep learning techniques for image recognition and image generation.

興味をお持ちの方は，相澤 hiroaki-aizawa {at} hiroshima-u.ac.jp までご連絡ください．

Members

研究室メンバー

Faculty

教員

  1. 相澤宏旭（准教授） [個人ページ]

Master's Students

博士課程前期学生

  1. 陳 茂森
  2. 李 微奇
  3. 米花 陽希
  4. 篠原 彰斗
  5. 久保田 晃市

Undergraduate Students

学部生

  1. 加藤 颯馬
  2. 渕上 塁以
  3. 藤中 春樹
  4. 高橋 拓巳

Research Students

研究生

  1. 呉 楽楽
  2. 弓 馳弘

Research

研究

2021年以前の論文，パターン認識研究室（栗田研究室）の研究，国内論文誌と国内会議の研究は 栗田多喜夫先生の個人ページ もしくは 相澤の個人ページ をご覧ください．
For publications before 2021, research conducted in the Pattern Recognition Laboratory under Professor Kurita, and domestic journal and conference papers, please visit Professor Takio Kurita's personal page or Aizawa's personal page.

  1. Renyi Attention Entropy for Patch Pruning teaser

    Hiroaki Aizawa*, Yuki Igaue*

    Rényi Attention Entropy for Patch Pruning

    Proc. International Conference on Pattern Recognition (ICPR), pp.xx-xx, 2026, accepted, (*equal contribution).

  2. Logit Mixture Outlier Exposure teaser

    Akito Shinohara, Kohei Fukuda, Hiroaki Aizawa

    Logit Mixture Outlier Exposure for Fine-grained Out-of-Distribution Detection

    Proc. International Conference on Digital Image Computing: Techniques and Applications (DICTA), pp.1-7, 2025, refereed.

  3. Flatness-aware Curriculum Learning teaser

    Hiroaki Aizawa, Yoshikazu Hayashi

    Flatness-aware Curriculum Learning via Adversarial Difficulty

    Proc. British Machine Vision Conference (BMVC), pp.1-12, 2025, refereed.

  4. Class-wise Flooding Regularization teaser

    Hiroaki Aizawa*, Yuta Naito*, Kohei Fukuda

    Class-wise Flooding Regularization for Imbalanced Image Classification

    Proc. Asian Conference on Pattern Recognition (ACPR), pp.177-187, 2025, refereed, (*equal contribution).

  5. Patch Pruning Strategy teaser

    Yuki Igaue, Hiroaki Aizawa

    Patch Pruning Strategy Based on Robust Statistical Measures of Attention Weight Diversity in Vision Transformers

    Proc. Asian Conference on Pattern Recognition (ACPR), pp.123-133, 2025, refereed.

  6. Learning Neural Velocity Fields teaser

    Sota Ito, Yoshikazu Hayashi, Hiroaki Aizawa, Kunihito Kato

    Learning Neural Velocity Fields from Dynamic 3D Scenes via Inter-Frame Consistency Loss

    Proc. International Joint Conference on Neural Networks (IJCNN), pp.1-8, 2025, refereed.

  7. Adaptive Out-of-Distribution Detection teaser

    Kohei Fukuda, Hiroaki Aizawa

    Adaptive Out-of-Distribution Detection with Coarse-to-Fine Grained Representation

    International Conference on Computer Vision Theory and Applications (VISAPP), pp.19-26, 2025

  8. Edge-Aware Ray Sampling teaser

    Sota Ito, Yoshikazu Hayashi, Hiroaki Aizawa, Kunihito Kato

    Learning Neural Velocity Fields from Dynamic 3D Scenes via Edge-Aware Ray Sampling

    International Conference on Computer Vision Theory and Applications (VISAPP), pp.699-706, 2025

  9. Few-Shot Visual Inspection Model teaser

    Shiryu Ueno, Yoshikazu Hayashi, Shunsuke Nakatsuka, Yusei Yamada, Hiroaki Aizawa, Kunihito Kato

    Vision-Language In-Context Learning Driven Few-Shot Visual Inspection Model

    International Conference on Computer Vision Theory and Applications (VISAPP), pp.253-260, 2025

  10. Explainability of Visual Inspection teaser

    Shiryu Ueno, Yoshikazu Hayashi, Shunsuke Nakatsuka, Hiroaki Aizawa, Kunihito Kato

    Boosting Explainability of Visual Inspection via Multi-Head Vision Language Model

    International Conference on Quality Control by Artificial Vision (QCAV), Vol.13737, p.1373702, 2025

  11. Taylor Outlier Exposure teaser

    Kohei Fukuda, Hiroaki Aizawa

    Taylor Outlier Exposure

    arXiv:2412.07219

  12. Spatial Locality in Vision Transformer teaser

    Yuki Igaue, Takio Kurita, Hiroaki Aizawa

    Incorporating Spatial Locality into Self-Attention for Training Vision Transformer on Small-Scale Datasets

    Proceedings of 27th International Conference on Pattern Recognition (ICPR), Lecture Notes in Computer Science (volume 15303), pp.259-274, 2024

  13. Classification of Carotid Plaque teaser

    Takeshi Yoshidomi, Shinji Kume, Hiroaki Aizawa, Akira Furui

    Classification of Carotid Plaque with Jellyfish Sign Through Convolutional and Recurrent Neural Networks Utilizing Plaque Surface Edges

    Proceedings of the 46th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC2024), pp.1-4, 2024

  14. Important Pixels Sampling teaser

    Kohei Fukuda, Takio Kurita, Hiroaki Aizawa

    Important Pixels Sampling for NeRF Training based on Edge Values and Squared Errors between the Ground Truth and the Estimated Colors

    International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) - Volume 2: VISAPP, pp.102-111, 2024

  15. Eigengrasp-Conditioned Neural Radiance Fields teaser

    Hiroaki Aizawa and Itoshi Naramura

    Eigengrasp-Conditioned Neural Radiance Fields

    IEEE Access, vol.11, pp.121629-121636, 2023

  16. Image inpainting teaser

    Yin Wang, Hiroaki Aizawa, Takio Kurita

    Image inpainting for large and irregular mask based on partial convolution and cross semantic attention

    Asian Conference on Pattern Recognition (ACPR), pp.115-128, 2023

  17. Few-shot NeRF-based View Synthesis teaser

    Sota Ito, Hiroaki Aizawa, Kunihito Kato

    Few-shot NeRF-based View Synthesis for Viewpoint-biased Camera Pose Estimation

    International Conference on Artificial Neural Networks (ICANN), pp.308-319, 2023

  18. Neural Radiance Fields with Regularizer teaser

    Kohei Fukuda, Takio Kurita, Hiroaki Aizawa

    Neural Radiance Fields with Regularizer Based on Differences of Neighboring Pixels

    International Joint Conference on Neural Networks (IJCNN), pp.1-7, 2023

  19. Facial Image Manipulation teaser

    Jiazhou Zheng, Hiroaki Aizawa, Takio Kurita

    Facial Image Manipulation via Discriminative Decomposition of Semantic Space

    International Joint Conference on Neural Networks (IJCNN), pp.1-8, 2023

  20. Parallax-based Imitation Learning teaser

    Yasuharu Niwa, Kunihito Kato, Hiroaki Aizawa, Yoshiyuki Hatta, Kazuaki Ito

    Parallax-based Imitation Learning with Human Intervention for Uncertain Insertion Tasks

    International Workshop on Frontiers of Computer Vision (IW-FCV), pp.271-285, 2023

  21. Frequency Perturbation Analysis teaser

    Yoshikazu Hayashi, Hiroaki Aizawa, Shunsuke Nakatsuka, Kunihito Kato

    Frequency Perturbation Analysis for Anomaly Detection using Fourier Heat Map

    International Conference on Quality Control by Artificial Vision (QCAV), Vol.12749, pp.73-79, 2023

  22. Additional Learning for BiGAN teaser

    Jiazhou Zheng, Hiroaki Aizawa, Takio Kurita

    Additional Learning for Joint Probability Distribution Matching in BiGAN

    International Conference on Neural Information Processing (ICONIP), pp.165-176, 2022

  23. Neural Density-Distance Fields teaser

    Itsuki Ueda, Yoshihiro Fukuhara, Hirokatsu Kataoka, Hiroaki Aizawa, Hidehiko Shishido, Itaru Kitahara

    Neural Density-Distance Fields

    European Conference on Computer Vision (ECCV), pp.53-68, 2022

Contact

連絡先

Address

宛先

〒739-8527
広島県東広島市鏡山1-4-1
広島大学 先進理工系科学研究科 情報科学プログラム 総合科学部C棟6階C619
相澤 宏旭

Office

居室

広島大学 東広島キャンパス 総合科学部C棟6階C619
TEL: 082-424-4655 (内線:4655)

Laboratory

研究室

広島大学 東広島キャンパス 総合科学部C棟7階C706