Bumsoo Kim

AI + Graphics Researcher

Which Ben will greet you next?!

Love collaborating at the intersection of art + tech!

Contact


Google Scholar
Loading...

Local Time (GMT+09:00)
Book appointment

Republic of Korea

Which Ben will greet you next?!

Contact

bumsookim00@gmail.com
Google Scholar

Local Time (GMT+09:00)
Book appointment

Republic of Korea


I'm actively seeking a research scientist position for alternative military service! for Fall 2026

I am a ML-aided visual computing researcher—computer vision, computer graphics, and generative AI. My academic footprints cross 2D/3D editing, multimodal generation, and neural rendering system for entertainment fields.




Selected Publications
Sounds That Shape: Audio-Driven 3D Mesh Generation with Attribute-Decoupled Score Distillation Sampling
Bumsoo Kim, Sanghyun Seo
ICASSP, 2026
Make VLM Recognize Visual Hallucination or Cartoon Character Image with Pose Information
Bumsoo Kim*, Wonseop Shin*, Kyuchul Lee, Yonghoon Jung, Sanghyun Seo
WACV, 2025
 Chatgpt Demo  HuggingFace Dataset
ToonAging: Face Re-Aging upon Artistic Portrait Style Transfer
Bumsoo Kim, Abdul Muqeet, Kyuchul Lee, Sanghyun Seo
CVPR AI4CC Workshop, 2024
Minecraft-ify: Minecraft Style Image Generation with Text-guided Image Editing for In-Game Application
Bumsoo Kim, Sanghyun Byun, Yonghoon Jung, Wonseop Shin, Sareer UI Amin, Sanghyun Seo
NeurIPS ML4CD Workshop (Oral), 2023

All (Chronological)
TBD

3D Data
Taming Gaussian Densification in Score Distillation Sampling for High-Fidelity Text-to-3D Generation
Bumsoo Kim, Junghun Ryu, Sanghyun Seo
CGI, 2026 (Under review)
 Code (Soon)
Sounds That Shape: Audio-Driven 3D Mesh Generation with Attribute-Decoupled Score Distillation Sampling
Bumsoo Kim, Sanghyun Seo
ICASSP, 2026
 Code (Soon)
MultiFloodSynth: Multi-Annotated Flood Synthetic Dataset Generation
YoonJe Kang*, Yonghoon Jung*, Wonseop Shin*, Bumsoo Kim*, Sanghyun Seo
AAAI GoodData Workshop (Oral), 2025

Photos
VideoTimeTravel: high-fidelity face re-aging diffusion models for production video
Bumsoo Kim, Yunyoung Nam, Sanghyun Seo
Visual Computer (IF:2.9), 2026
ToonAging: Face Re-Aging upon Artistic Portrait Style Transfer
Bumsoo Kim, Abdul Muqeet, Kyuchul Lee, Sanghyun Seo
CVPR AI4CC Workshop, 2024
Video Face Re-Aging: Toward Temporally Consistent Face Re-Aging
Abdul Muqeet*, Kyuchul Lee*, Bumsoo Kim, Yohan Hong, Hyungrae Lee, Woonggon Kim, KwangHee Lee
arXiv, 2023

Arts & Gaming
How Generative Models Serve as Visual Artists for Creative Interpolation
Bumsoo Kim, Yoonje Kang, Sanghyun Seo
CVPR Workshop, 2026 (Under review)
Learning Pose‑to‑Semantic for 3D Anamorphic Art based on 2D Diffusion Guidance
MS Thesis, 2026
Make VLM Recognize Visual Hallucination or Cartoon Character Image with Pose Information
Bumsoo Kim*, Wonseop Shin*, Kyuchul Lee, Yonghoon Jung, Sanghyun Seo
WACV, 2025
 Chatgpt Demo  HuggingFace Dataset
Minecraft-ify: Minecraft Style Image Generation with Text-guided Image Editing for In-Game Application
Bumsoo Kim, Sanghyun Byun, Yonghoon Jung, Wonseop Shin, Sareer UI Amin, Sanghyun Seo
NeurIPS ML4CD Workshop (Oral, Top 8.7%), 2023

Real-world
EfficientNetV2-based dynamic gesture recognition using transformed scalogram from triaxial acceleration signal
Bumsoo Kim, Sanghyun Seo
JCDE (IF:4.9), 2023
Intelligent Digital Human Agent Service With Deep Learning Based-Face Recognition
Bumsoo Kim, Sanghyun Seo
IEEE Access (IF:3.9), 2022

Biography (2026)

Bumsoo Kim received the BSc and MSc in art and technology from Chung-Ang University, South Korea, in 2023, 2026. He worked for machine learning algorithm and its industrial system with ETRI, VIVE STUDIOS, and Boeing Research & Technology, South Korea. His research areas include generative models and their applications on computer graphics.

Academic lineage

Shadow-Art Object Creation Tool | MS Thesis (2026)

3D generative models. TBD



Dynamic Projection Mapping | BS Graduation Exhibition (2023)

Computer vision technics + RGB-D camera
I developed a vision system that captures movable objects in real-time. To do that, I integrate RGB-D camera and video projector with GPU-accelerated Just-In-Time compiler on python interface. For visual contents, I optimized YOLACT model with arbitrary neural style transfer in edge devices.


My Official Code Implementation

TBD

My Unofficial Code Implementation

 EffectiveResolutionEstimation (2024)

 PixelCycleGAN (2023)

 CycleGAN (2023)


My Stable Maintenance Version from Official Repo

 FADING (2024~)