1. X
  2. Cheng Lin
Log inSign up
Cheng Lin
194 posts
Cheng Lin profile banner
user avatar

Cheng Lin

@_cheng_lin
Associate Professor @MUST | Prev Research Scientist @Tencent&MiHoYo | Organizer @AnySyn3D 3D Vision, 3D AIGC, Geometric Modeling, Graphics
clinplayer.github.io
Joined March 2020
333
Following
922
Followers
RepliesRepliesMediaMedia
  • Pinned
    user avatar
    Cheng Lin
    @_cheng_lin
    Jul 29, 2025
    🔎Excited to share our PDT: Point Distribution Transformation with Diffusion Models (SIGGRAPH2025)! 💡While autoregressive models are now widely adopted for predicting structures, we find diffusion models can also reveal high-level structures by transforming point distributions!
    Image
    00:00
  • user avatar
    Cheng Lin
    @_cheng_lin
    Jul 2
    Thanks to Zhiyang for providing the GPU version. Coverage Axis was our very first collaborative paper from four years ago. The method is simple yet robust, offering an innovative solution for shape skeletonization.
    user avatar
    Zhiyang (Frank) Dou
    @frankzydou
    Jul 1
    If you work on shape skeletonization/medial axis/point sampling, you may want to check this out. We made Coverage Axis a GPU version to accelerate it. The point-selection step is a set-cover problem, and we solve it exactly on the GPU, so you get the same optimal selected
    Image
    Image
    Image
  • user avatar
    Cheng Lin
    @_cheng_lin
    Apr 28
    Great Work!
    user avatar
    Zhiyang (Frank) Dou
    @frankzydou
    Apr 27
    SAD: Soft Anisotropic Diagrams for Differentiable Image Representation has been accepted by #SIGGRAPH2026 Check it out, and huge congrats to Lucky! @Luckyballa #SAD represents an image as a soft, anisotropic, differentiable diagram over learnable sites. Each pixel is modeled
  • user avatar
    Cheng Lin
    @_cheng_lin
    Mar 13
    ✨Introducing our research product FizzReel🎥 — a video re-creation tool that uses AI to understand the creative intent behind your reference, from shot design and VFX to lighting💡. Combined with powerful models like Seedance 2.0🔥, it's the future of video creation! 🚀
    Image
    00:00
  • user avatar
    Cheng Lin
    @_cheng_lin
    Dec 4, 2025
    Welcome to check out our spotlight paper at NeurIPS2025!🌟
    user avatar
    Zhiyang (Frank) Dou
    @frankzydou
    Dec 3, 2025
    Please check out paper #MOSPA "🎧Human Motion Generation Driven by Spatial Audio” at #NeurIPS2025 (🌟Spotlight)! 😊We have released our dataset and models : ) 💡The paper tackles the challenge of spatial-audio-driven human motion generation, enabling virtual humans to respond
    Image

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Advertisement
Advertisement