From a Single Image to a Physically Realistic 4D Video in One Minute

PhysGM, a CVPR 2026 paper by Beijing Institute of Technology and Li Auto, transforms a single static image into a high‑fidelity 4D video that obeys real‑world physics in under a minute, using a dual‑decoder transformer, DPO alignment, and a newly built 50k‑item PhysAssets dataset, outperforming prior methods in speed and quality.

3D Gaussian SplattingCVPR 2026Direct Preference Optimization

0 likes · 7 min read

From a Single Image to a Physically Realistic 4D Video in One Minute

Kuaishou Large Model

Jul 29, 2022 · Fundamentals

How Automatic Quantization Slashes Memory Use in High‑Resolution Physical Simulations

This article explains how researchers applied quantization techniques to high‑resolution physical simulations, enabling over 50% memory reduction without noticeable visual loss, by modeling error propagation, using constrained optimization, and introducing dithering, with results demonstrated on GPU‑based smoke, fluid, and elastic body simulations.

GPU memory optimizationPhysical SimulationSIGGRAPH

0 likes · 6 min read

How Automatic Quantization Slashes Memory Use in High‑Resolution Physical Simulations

Kuaishou Tech

Jul 14, 2021 · Fundamentals

QuanTaichi: A Physical Compiler for Automatic Quantization of High‑Precision Simulations

QuanTaichi, built on the Taichi language, introduces custom numeric types, bit‑struct adapters, and compiler optimizations that dramatically reduce memory and bandwidth for particle‑based physics simulations, enabling high‑precision GPU rendering on a single card and even on mobile devices.

GPU simulationGraphicsPhysical Simulation

0 likes · 13 min read

QuanTaichi: A Physical Compiler for Automatic Quantization of High‑Precision Simulations

21CTO

Jan 13, 2020 · Fundamentals

How 99 Lines of Code Let You Create Frozen‑Style Magic with Taichi

Taichi, a new CG programming language created by MIT PhD student Hu Yuanming, enables high‑performance physical simulation and differentiable programming with just a few dozen lines of code, dramatically outperforming TensorFlow and PyTorch, and has been showcased in SIGGRAPH, ICRA, ICLR papers and real‑time demos.

Differentiable ProgrammingMLSM‑MPMPhysical Simulation

0 likes · 10 min read

How 99 Lines of Code Let You Create Frozen‑Style Magic with Taichi