ARC Lab, Tencent PCG

[CVPR 2026] Spatio-Temporal Autoregressive 4K 360° Video Generation from Perspective Video

Track4World: Feedforward World-centric Dense 3D Tracking of All Pixels

InstantMesh: Efficient 3D Mesh Generation from a Single Image with Sparse-view Large Reconstruction Models

VerseCrafter: Dynamic Realistic Video World Model with 4D Geometric Control

[ICCV 2023] Consistent Image Synthesis and Editing

[ICLR 2026] Streamlining Cartoon Production with Generative Post-Keyframing

[CVPR 2026] MotionCrafter: Dense Geometry and Motion Reconstruction with a 4D VAE

T2I-Adapter

PhotoMaker [CVPR 2024]

AudioStory: Generating Long-Form Narrative Audio with Large Language Models

GFPGAN aims at developing Practical Algorithms for Real-world Face Restoration.

SEED-Voken: A Series of Powerful Visual Tokenizers

[ICLR 2026] Official Repo for Rolling Forcing: Autoregressive Long Video Diffusion in Real Time

SEED-Story: Multimodal Long Story Generation with Large Language Model

[Arxiv'25] IC-Custom: Diverse Image Customization via In-Context Learning

[CVPR 2026] TimeLens: Rethinking Video Temporal Grounding with Multimodal LLMs

[ICLR 2026] GenCompositor: Generative Video Compositing with Diffusion Transformer

Structuring Hour-Long Videos into Navigable Chapters and Hierarchical Summaries

[SIGGRAPH2025] Official repo for paper "Any-length Video Inpainting and Editing with Plug-and-Play Context Control"

Code release of our paper "DI-PCG: Diffusion-based Efficient Inverse Procedural Content Generation for High-quality 3D Asset Creation".

[CVPR 2025] Official code of "DiTCtrl: Exploring Attention Control in Multi-Modal Diffusion Transformer for Tuning-Free Multi-Prompt Longer Video Generation"

Official code of SmartEdit [CVPR-2024 Highlight]

TokLIP: Marry Visual Tokens to CLIP for Multimodal Comprehension and Generation

The official implementation of paper "ColorFlow: Retrieval-Augmented Image Sequence Colorization". ColorFlow：基于检索增强的图像序列上色

HOSNeRF: Dynamic Human-Object-Scene Neural Radiance Fields from a Single Video

A framework to convert any 2D videos to immersive stereoscopic 3D

Structured Video Comprehension of Real-World Shorts

Video-Holmes: Can MLLM Think Like Holmes for Complex Video Reasoning?