👋 Hi, I'm Seongchan Kim (김성찬)

🎥 Video Generation & Multimodal Large Language Models (MLLM)
🧑‍💻 Integrated M.S./Ph.D. @CVLAB in KAIST AI

I design next-generation video generation models and build evaluation frameworks for understanding and improving video diffusion models.
Currently exploring interaction-aware video generation and multimodal understanding of videos.

---

🧪 Research Highlights

• 🎬 Video Generation & Evaluation — Improving interaction fidelity and multi-instance understanding in video diffusion transformers
• 🧩 Video Object Segmentation (VOS) — Multi-granularity & referring VOS with language and temporal reasoning
• 🧠 MLLM for Video — Leveraging multimodal large language models to better understand and describe video content

---

📝 Publications

• Self-Evolving Neural Radiance Fields
  Wild3D Workshop @ ICCV 2025
  🔗 Project Page

• MUG-VOS: Multi-Granularity Video Object Segmentation
  AAAI 2025
  🔗 Project Page

• Referring Video Object Segmentation via Language Aligned Track Selection
  arXiv 2025
  🔗 Project Page

• InterRVOS: Interaction-aware Referring Video Object Segmentation
  Under review at AAAI 2026
  🔗 Project Page

• MATRIX: Mask Track Alignment for Interaction-Aware Video Generation
  Under review at ICLR 2026

---

🌎 Links

• 🌐 Google Scholar
• 💼 LinkedIn
• 🐦 X (Twitter)
• 🏠 Personal Website

✨ “Understanding the World through Video and Multimodalities.”

_{🔄 Last updated: 2025년 9월 28일 | 💻 Made with ❤️ by Deep Overflow}