Yichun Shi
Scholar

Yichun Shi

Google Scholar ID: RXZChV0AAAAJ
ByteDance
Computer VisionMachine Learning
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
3,467
 
H-index
25
 
i10-index
31
 
Publications
20
 
Co-authors
0
 
Contact
CVOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
13 items
Towards Robust Sequential Decomposition for Complex Image Editing
2026
Cited
0
MMCORE: MultiModal COnnection with Representation Aligned Latent Embeddings
2026
Cited
0
VTok: A Unified Video Tokenizer with Decoupled Spatial-Temporal Latents
2026
Cited
0
RapidMV: Leveraging Spatio-Angular Representations for Efficient and Consistent Text-to-Multi-View Synthesis
2025
Cited
0
Seedream 4.0: Toward Next-generation Multimodal Image Generation
2025
Cited
0
VINCIE: Unlocking In-context Image Editing from Video
2025
Cited
0
SeedEdit 3.0: Fast and High-Quality Generative Image Editing
2025
Cited
0
ByteMorph: Benchmarking Instruction-Guided Image Editing with Non-Rigid Motions
2025
Cited
0
Resume
Academic Achievements
  • SeedEdit: Align Image Re-generation to Image Editing
  • MVDream: Multi-view Diffusion for 3D Generation (ICLR, 2024)
  • PAniC-3D: Stylized Single-view 3D Reconstruction from Portraits of Anime Characters (CVPR, 2023)
  • PanoHead: Geometry-Aware 3D Full-Head Synthesis in 360 (CVPR, 2023)
  • OmniAvatar: Geometry-Guided Controllable 3D Head Synthesis (CVPR, 2023)
  • IDE-3D: Interactive Disentangled Editing for High-Resolution 3D-aware Portrait Synthesis (ACM Trans. on Graphics, SIGGRAPH Asia, 2022)
  • SemanticStyleGAN: Learning Compositonal Generative Priors for Controllable Image Synthesis and Editing (CVPR, 2022)
  • Lifting 2D StyleGAN for 3D-aware Face Generation (CVPR, 2021)
  • Probabilistic Face Embeddings (ICCV, 2019)
  • WarpGAN: Automatic Caricature Generation (CVPR, 2019)
Background
  • Currently a research scientist at ByteDance Inc (Tiktok). Research interests include Generative Models and Neural Rendering, particularly image, video, and 3D generation.