Tencent's VerseCrafter: A Dynamic Video World Model with 4D Geometric Control
tom_doerr · x · 2026-08-10
Tencent ARC Lab, in collaboration with Fudan University and HKU, has open-sourced VerseCrafter, a dynamic realistic video world model.
The project features explicit 4D geometric control over camera and multi-object motion to generate high-fidelity dynamic videos. Code and checkpoints are now available on GitHub.
More from Multimodal
- Visual Comparison: H3 Int8 ConvRot vs W4A8_mixed Quantization — Devajyoti1231 · 2026-08-10
- Grok Imagine 2.0 Ships, Jumping to #2 Globally in Image Generation — eyishazyer · 2026-08-10
- SEEDANCE 2 Generates Sci-Fi Action Trailer for Zero-Gravity F1 Racers — tess-tipple · 2026-08-10
- Workflow and Prompts for Generating Celebrity Characters with Minimax H3 Locally — Fabulous-Snow4366 · 2026-08-10
- Nous Portal expands multimodal library with new image and video generation models — Teknium · 2026-08-10
- SweepGrid: Sweep ComfyUI Parameters and Generate Comparison Videos — YouKilledApollo · 2026-08-10