13 papers
Biological Congruence: Evaluating Microscopy Generation Across Representations
Duway Nicolas Lesmes-Leon, Shumpei Takezaki, Fabian Schmeisser +5
Generative models increasingly produce realistic microscopy images, yet conventional evaluation primarily characterizes visual fidelity and distributional coverage rather than the…
Cell Instance Segmentation via Multi-Task Image-to-Image Schrödinger Bridge
Hayato Inoue, Shota Harada, Shumpei Takezaki +1
Existing cell instance segmentation pipelines typically combine deterministic predictions with post-processing, which imposes limited explicit constraints on the global structure o…
SCoRe: Clean Image Generation from Diffusion Models Trained on Noisy Images
Yuta Matsuzaki, Seiichi Uchida, Shumpei Takezaki
Diffusion models trained on noisy datasets often reproduce high-frequency training artifacts, significantly degrading generation quality. To address this, we propose SCoRe (Spectra…
VesselFusion: Diffusion Models for Vessel Centerline Extraction from 3D CT Images
Soichi Mita, Shumpei Takezaki, Ryoma Bise
Vessel centerline extraction from 3D CT images is an important task because it reduces annotation effort to build a model that estimates a vessel structure. It is challenging to es…
Few-Part-Shot Font Generation
Masaki Akiba, Shumpei Takezaki, Daichi Haraguchi +1
This paper proposes a novel model of few-part-shot font generation, which designs an entire font based on a set of partial design elements, i.e., partial shapes. Unlike conventiona…
NoiseCutMix: A Novel Data Augmentation Approach by Mixing Estimated Noise in Diffusion Models
Shumpei Takezaki, Ryoma Bise, Shinnosuke Matsuo
In this study, we propose a novel data augmentation method that introduces the concept of CutMix into the generation process of diffusion models, thereby exploiting both the abilit…
Inverse Scene Text Removal
Takumi Yoshimatsu, Shumpei Takezaki, Seiichi Uchida
Scene text removal (STR) aims to erase textual elements from images. It was originally intended for removing privacy-sensitiveor undesired texts from natural scene images, but is n…
Guidance-base Diffusion Models for Improving Photoacoustic Image Quality
Tatsuhiro Eguchi, Shumpei Takezaki, Mihoko Shimano +2
Photoacoustic(PA) imaging is a non-destructive and non-invasive technology for visualizing minute blood vessel structures in the body using ultrasonic sensors. In PA imaging, the i…
Self-Relaxed Joint Training: Sample Selection for Severity Estimation with Ordinal Noisy Labels
Shumpei Takezaki, Kiyohito Tanaka, Seiichi Uchida
Severity level estimation is a crucial task in medical image diagnosis. However, accurately assigning severity class labels to individual images is very costly and challenging. Con…
Cross-Domain Image Conversion by CycleDM
Sho Shimotsumagari, Shumpei Takezaki, Daichi Haraguchi +1
The purpose of this paper is to enable the conversion between machine-printed character images (i.e., font images) and handwritten character images through machine learning. For th…
An Ordinal Diffusion Model for Generating Medical Images with Different Severity Levels
Shumpei Takezaki, Seiichi Uchida
Diffusion models have recently been used for medical image generation because of their high image quality. In this study, we focus on generating medical images with ordinal classes…
Font Style Interpolation with Diffusion Models
Tetta Kondo, Shumpei Takezaki, Daichi Haraguchi +1
Fonts have huge variations in their styles and give readers different impressions. Therefore, generating new fonts is worthy of giving new impressions to readers. In this paper, we…
Disease Severity Regression with Continuous Data Augmentation
Shumpei Takezaki, Kiyohito Tanaka, Seiichi Uchida +1
Disease severity regression by a convolutional neural network (CNN) for medical images requires a sufficient number of image samples labeled with severity levels. Conditional gener…