Takumi Karasawa — accepted AI-conference papers

2025

Multi-Point Positional Insertion Tuning for Small Object Detection

Kanoko Goto, Takumi Karasawa, Takumi Hirose, Rei Kawakami, Nakamasa Inoue

ICASSP 2025accepted

Small object detection aims to localize and classify small objects within images. With recent advances in large-scale vision-language pretraining, finetuning pretrained object detection models has emerged as a promising approach. However, finetuning large models is computationally and memory expensi…

Cited by 0 Source ↗Scholar ↗

2017

MFNet: Towards real-time semantic segmentation for autonomous vehicles with multi-spectral scenes

Qishen Ha, Kohei Watanabe, Takumi Karasawa, Yoshitaka Ushiku, Tatsuya Harada

IROS 2017poster

This work addresses the semantic segmentation of images of street scenes for autonomous vehicles based on a new RGB-Thermal dataset, which is also introduced in this paper. An increasing interest in self-driving vehicles has brought the adaptation of semantic segmentation to self-driving systems. Ho…

Cited by 617 Source ↗Scholar ↗