SIFTer: Self-improving Synthetic Datasets for Pre-training Classification Models, Ryo Hayamizu, Shota Nakamura, Sora Takashima, Hirokatsu Kataoka, Ikuro Sato, Nakamasa Inoue, Rio Yokota, CVPR 2024 Workshop. [Paper] [OpenReview] [Workshop]

Primitive Geometry Segment Pre-training for 3D Medical Image Segmentation, Ryu Tadokoro, Ryosuke Yamada, Kodai Nakashima, Ryo Nakamura, Hirokatsu Kataoka, BMVC 2023 Best Industry Paper Finalist. [Paper] [Code] [Video]

SegRCDB: Semantic Segmentation via Formula-Driven Supervised Learning, Risa Shinoda, Ryo Hayamizu, Kodai Nakashima, Nakamasa Inoue, Rio Yokota, Hirokatsu Kataoka, ICCV 2023. [Paper] [Code] [Models]

Pre-training Vision Transformers with Very Limited Synthesized Images, Ryo Nakamura, Hirokatsu Kataoka, Sora Takashima, Edgar Josafat Martinez Noriega, Rio Yokota, Nakamasa Inoue, ICCV 2023. [Paper] [Project] [Code] [Dataset] [Poster]

Does Formula-Driven Supervised Learning Work on Small Datasets?, Kodai Nakashima, Hirokatsu Kataoka, Yutaka Satoh, IEEE Access 2023. [Paper]

Visual Atoms: Pre-training Vision Transformers with Sinusoidal Waves, Sora Takashima, Ryo Hayamizu, Nakamasa Inoue, Hirokatsu Kataoka, Rio Yokota, CVPR 2023. [Paper] [Project] [Code]

Pre-training Auto-generated Volumetric Shapes for 3D Medical Image Segmentation, Ryu Tadokoro, Ryosuke Yamada, Hirokatsu Kataoka, CVPR 2023 Workshop.

Point Cloud Pre-training with Natural 3D Structures, Ryosuke Yamada, Hirokatsu Kataoka, Naoya Chiba, Yukiyasu Domae, Tetsuya Ogata, CVPR 2022. [Paper] [Project] [Code]

Replacing Labeled Real-Image Datasets with Auto-Generated Contours, Hirokatsu Kataoka, Ryo Hayamizu, Ryosuke Yamada, Kodai Nakashima, Sora Takashima, Xinyu Zhang, Edgar Josafat Martinez-Noriega, Nakamasa Inoue, Rio Yokota, CVPR 2022. [Paper] [Project] [Code] [Oral] [Poster]

Pre-training without Natural Images
Hirokatsu Kataoka, Kazushige Okayasu, Asato Matsumoto, Eisuke Yamagata, Ryosuke Yamada, Nakamasa Inoue, Akio Nakamura, Yutaka Satoh, IJCV 2022. [Paper] [Project] [Code]

Can Vision Transformers Learn without Natural Images?
Kodai Nakashima, Hirokatsu Kataoka, Asato Matsumoto, Kenji Iwata, Nakamasa Inoue, AAAI 2022. [Paper] [Project] [Code] [Dataset]

Spatiotemporal Initialization for 3D CNNs with Generated Motion Patterns
Hirokatsu Kataoka, Eisuke Yamagata, Kensho Hara, Ryusuke Hayashi, Nakamasa Inoue, WACV 2022. [Paper] [Project]

Formula-driven Supervised Learning with Recursive Tiling Patterns
Hirokatsu Kataoka, Asato Matsumoto, Eisuke Yamagata, Ryosuke Yamada, Nakamasa Inoue, Yutaka Satoh, ICCV 2021 Workshop. [Paper] [Project]

MV-FractalDB: Formula-driven Supervised Learning for Multi-view Image Recognition
Ryosuke Yamada, Ryo Takahashi, Ryota Suzuki, Akio Nakamura, Yusuke Yoshiyasu, Ryusuke Sagawa, Hirokatsu Kataoka, IROS 2021. [Paper] [Project]

Initialization Using Perlin Noise for Training Networks with a Limited Amount of Data
Nakamasa Inoue, Eisuke Yamagata, Hirokatsu Kataoka, ICPR 2020. [Paper] [YouTube]

Pre-training without Natural Images
Hirokatsu Kataoka, Kazushige Okayasu, Asato Matsumoto, Eisuke Yamagata, Ryosuke Yamada, Nakamasa Inoue, Akio Nakamura, Yutaka Satoh, ACCV 2020 Best Paper Honorable Mention Award.
[Paper] [Project] [Code] [Oral] [Poster] [Supp. Mat.]

CVPR 2024 Workshop on Representation Learning with Very Limited Images: Zero-shot, Unsupervised, and Synthetic Learning in the Era of Big Models [Link]

ICCV 2023 Workshop on Representation Learning with Very Limited Images: The potential of self-, synthetic-, and formula-supervision [Link]

Pre-training without Natural Images
Hirokatsu Kataoka
IW-FCV 2023
[Link] [Slide]

Pre-training without Natural Images
Hirokatsu Kataoka
MIT
[Slide]

自然法則に基づく深層学習
Hirokatsu Kataoka
NVIDIA HPC Week - HPC + Machine Learning
[Link] [Slide]

限られたデータからの深層学習
Nakamasa Inoue
MIRU 2021
[Link] [Slide]

Pre-training without Natural Images
Hirokatsu Kataoka
SNL 2021
[Link] [Slide]