3d Avatar Research and Development - Pico Perception - San Jose

ByteDance ByteDance · Big Tech · San Jose, CA · R&D

Research and development role focusing on 3D Avatar generative models, involving 3D geometry, texturing, human reconstruction, and animation techniques. Requires expertise in generative modeling for 3D/4D reconstruction/generation, with a Master's degree or above and proficiency in deep learning frameworks.

What you'd actually do

  1. Participate in research and development tasks to explore new frontiers in 3D Avatar generative models and deep dive into advanced 3D geometry, texturing, human reconstruction, and animation techniques.
  2. Contribute to project discussions, offering innovative ideas for 3D modeling and avatar generation/animation.
  3. Collaborate closely with XFN teams to refine 3D Avatar Generative models in business operations and transform advanced technology into PICO products.
  4. Document research findings clearly and contribute to research publications, patents, and presentations.

Skills

Required

  • Master's degree or above in Computer Vision, Computer Graphics, Mathematics, Game Design, Digital Arts, Animation, or related fields
  • Expertise in 3D modeling, texturing, and avatar generation
  • Research and practical experience in 3D face/head/hand/full-body modeling, editing, and driving
  • Experience with Explicit and implicit representations of the face/head/hand/full-body (e.g., NeRF, 3DGS, Codec Avatar)
  • Experience with Generative modeling (e.g., Diffusion or Transformer models) for 3D/4D reconstruction/generation
  • Algorithm design and implementation
  • Python/C/C++ programming
  • Deep learning frameworks (preferably PyTorch)

Nice to have

  • Strong research ability with publications in accredited computer vision and graphics conferences and journals
  • Good communication skills
  • Self-motivated
  • Thirst for knowledge, curiosity, and initiative
  • Passion for tackling technical challenges

What the JD emphasized

  • publications in accredited computer vision and graphics conferences and journals

Other signals

  • 3D Avatar generative models
  • 3D geometry, texturing, human reconstruction, and animation techniques
  • Generative modeling (e.g., Diffusion or Transformer models) for 3D/4D reconstruction/generation