Skip to content
View Yusameki's full-sized avatar
🌧️
rainy
🌧️
rainy
  • Keio University
  • China

Block or report Yusameki

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Yusameki/README.md

こんにちは、Meki です 🌧️

Meki Yusa

English · 日本語 · 中文

Computer vision, 3D perception, and the occasional write-up With Aina.


typing

Profile views

GitHub Streak


What I'm up to

  • 🔬 3D vision / hand-eye calibration — getting robot arms and 3D scanners to know each other
  • ☁️ Time-series forecasting — pushing satellite cloud imagery through ConvLSTM
  • 🌍 Exploring the world with Aina — local generative tooling on a 12 GB box, from images toward animation, music, and whatever comes next

A few repositories

Repository What it is
Local_Image_with_Aina 🆕 SDXL parameter baselines, a from-scratch MOSS-Image deployment on a 12 GB card, and a LoRA training log — the first step of Project_Aina
dsh-local-fullstack-256k-comfyui A 64GB RAM / 12GB VRAM consumer box turned into a fully local Agent workstation — Gyro-M at 256K context, 2.1K tool budget, persistent memory, on-demand engine start/stop, and ComfyUI image generation time-shared on the same GPU
eye-to-hand-calibration Hand-eye calibration in HALCON, Photoneo PhoXi + calibration sphere, with full point cloud and pose data
cloud Keras ConvLSTM forecasting of infrared imagery

Tools I use

Python PyTorch TensorFlow OpenCV Docker Linux


Mostly in the study, occasionally here ( ¯꒳¯ )

Pinned Loading

  1. dsh-local-fullstack-256k-comfyui dsh-local-fullstack-256k-comfyui Public

    A 64 GB RAM / 12 GB VRAM consumer box running a fully local Agent stack — Gyro-M at 256K context, 2.1K tool budget, persistent memory, on-demand engine start/stop, and time-shared ComfyUI image gen…

    Python 1

  2. Local_Image_with_Aina Local_Image_with_Aina Public

    SDXL parameter baselines · MOSS-Image local deployment · LoRA training log

    Python

  3. eye-to-hand-calibration eye-to-hand-calibration Public

    Eye-to-hand calibration with HALCON + Photoneo PhoXi 3D scanner.

  4. cloud cloud Public

    Keras ConvLSTM for Himawari-8 infrared satellite cloud imagery forecasting.

    Python