Xinhua Cheng

SPEED. [ICLR'26] SPEED: Scalable, Precise, and Efficient Concept Erasure for Diffusion Models

R-4B. The official repository of "R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Integration"

tinyworlds. A minimal implementation of DeepMind's Genie world model

LLMs-from-scratch. Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

llm_interview_note. 主要记录大语言大模型（LLMs） 算法（应用）工程师相关的知识及面试题

Self-Forcing. Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)

Matrix-Game. Matrix-Game 3.0: Real-Time and Streaming Interactive World Model with Long-Horizon Memory

OmniDrag. [IJCV 2025] OmniDrag: Enabling Motion Control for Omnidirectional Image-to-Video Generation

DanceGRPO. An official implementation of DanceGRPO: Unleashing GRPO on Visual Generation

UniWorld. UniWorld: High-Resolution Semantic Encoders for Unified Visual Understanding and Generation

Spark-Wan. Spark-Wan: Few-Step Video Diffusion Transformer via Hybrid Adversarial Post-Training

Q-Insight. Q-Insight Family: Q-Insight, VQ-Insight and RALI (NeurIPS 2025 Spotlight, AAAI 2026 Oral, and ICLR 2026 Oral)

ReCamMaster. [ICCV'25 Best Paper Finalist] ReCamMaster: Camera-Controlled Generative Rendering from A Single Video

VACE. [ICCV 2025] Official implementations for paper: VACE: All-in-One Video Creation and Editing

Phantom. Phantom: Subject-Consistent Video Generation via Cross-Modal Alignment

UniAnimate-DiT. UniAnimate-DiT: Human Image Animation with Large-Scale Video Diffusion Transformer

HoloTime. [ACM MM 2025] HoloTime: Taming Video Diffusion Models for Panoramic 4D Scene Generation

genwarp. Python

SkyReels-V2. SkyReels-V2: Infinite-length Film Generative model

VideoX-Fun. 📹 A more flexible framework that can generate videos at any resolution and creates videos from images.

DiffSynth-Studio. Enjoy the magic of Diffusion models!

Wan2.1. Wan: Open and Advanced Large-Scale Video Generative Models

CogKit. Finetuning and inference tools for the CogView4 and CogVideoX model series.

Depth-Anywhere. Python

pkuthss. LaTeX template for dissertations in Peking University

awesome-resume-for-chinese. :page_facing_up: 适合中文的简历模板收集（LaTeX，HTML/JS and so on）由 @hoochanlon 维护

vggt. [CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

TrajectoryCrafter. [ICCV 2025, Oral] TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Models

Depth-Anything-V2. [NeurIPS 2024] Depth Anything V2. A More Capable Foundation Model for Monocular Depth Estimation

Video-Depth-Anything. [CVPR 2025 Highlight] Video Depth Anything: Consistent Depth Estimation for Super-Long Videos

metric_depth_video_toolbox. Python tools for rendering, viewing and generating metric 3D depth videos. Tools for recovering and exporting camera pose and 3D geometry to popular formats as well as tools for projecting depthvideo in to side by side stereo(sbs) 3D stereo video.

MoGe. [CVPR'25 Oral] MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision

Imagine360. [Neurips 2025] Imagine360: Immersive 360 Video Generation from Perspective Anchor

SpaTracker. [CVPR 2024 Highlight] Official PyTorch implementation of SpatialTracker: Tracking Any 2D Pixels in 3D Space

DiffusionAsShader. [SIGGRAPH 2025] Diffusion as Shader: 3D-aware Video Diffusion for Versatile Video Generation Control

TrajectoryAttention. [ICLR 2025] Trajectory Attention For Fine-grained Video Motion Control

360-1M. Python

LuminaBrush. Illumination Drawing Tools for Text-to-Image Diffusion Models

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

Emu3. Next-Token Prediction is All You Need

Ruyi-Models. Python

CogVideo. text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

time_reversal. Official repo of "Time Reversal Fusion" (ECCV2024)

HunyuanVideo. HunyuanVideo: A Systematic Framework For Large Video Generation Model

ConsisID. [CVPR 2025 Highlight🔥] Identity-Preserving Text-to-Video Generation by Frequency Decomposition

DiffMorpher. Official Code for DiffMorpher: Unleashing the Capability of Diffusion Models for Image Morphing (CVPR 2024)

pytorch-image-models. The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

kohya_ss. Python

TVG. code for "TVG: A Training-free Transition Video Generation Method with Diffusion Models"

pidinet. Code for the ICCV 2021 paper "Pixel Difference Networks for Efficient Edge Detection" (Oral).

ResVR. [ACM MM 24 Best Paper Nomination] ResVR: Joint Rescaling and Viewport Rendering of Omnidirectional Images

ViewCrafter. [TPAMI 2025] ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesis

svd_keyframe_interpolation. Python

Kolors. Kolors Team

SPO. [CVPR 2025] Aesthetic Post-Training Diffusion Models from Generic Preferences with Step-by-step Preference Optimization

AnyText. Official implementation code of the paper <AnyText: Multilingual Visual Text Generation And Editing>

D3C2-Net. [TCSVT 2024] D3C2-Net: Dual-Domain Deep Convolutional Coding Network for Compressive Sensing

bidiff. [CVPR'24] Text-to-3D Generation with Bidirectional Diffusion using both 2D and 3D priors

interactive3d. [CVPR'24] Interactive3D: Create What You Want by Interactive 3D Generation

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

Awesome-Image-Harmonization. A curated list of papers, code and resources pertaining to image harmonization.

spad. Code for SPAD : Spatially Aware Multiview Diffusers, CVPR 2024

LN3Diff. [ECCV-2024] LN3Diff creates high-quality 3D object mesh from text within 8 V100-SECONDS.

GRM. Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generation

CRM. [ECCV 2024] Single Image to 3D Textured Mesh in 10 seconds with Convolutional Reconstruction Model.

Long-CLIP. [ECCV 2024] official code for "Long-CLIP: Unlocking the Long-Text Capability of CLIP"

generative-models. Generative Models by Stability AI

MVControl. [3DV-2025] Official implementation of "Controllable Text-to-3D Generation via Surface-Aligned Gaussian Splatting"

PixArt-alpha. PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis

Make-Your-3D. [ECCV 2024] Make-Your-3D: Fast and Consistent Subject-Driven 3D Content Generation

data-juicer. Data processing for and with foundation models! 🍎 🍋 🌽 ➡️ ➡️🍸 🍹 🍷

ComfyUI. The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.

Open-Sora-Plan. This project aim to reproduce Sora (Open AI T2V model), we wish the open source community contribute to this project.

FiT. [ICML 2024 Spotlight] FiT: Flexible Vision Transformer for Diffusion Model

ECDFormer. 【Nature Computational Science 2025🔥】Deep peak property learning for efficient chiral molecules ECD spectra prediction

DreamPropeller. Python

HeadStudio. [ECCV 2024] HeadStudio: Text to Animatable Head Avatars with 3D Gaussian Splatting.

GALA3D. [ICML 2024] GALA3D: Towards Text-to-3D Complex Scene Generation via Layout-guided Generative Gaussian Splatting

MIGC. [CVPR 2024 Highlight] MIGC and [TPAMI 2024] MIGC++ (Official Implementation)

LGM. [ECCV 2024 Oral] LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation.

Dynamic3DGaussians. Python

threestudio-gaussiandreamer. GaussianDreamer extension of threestudio.

TriplaneGaussian. TriplaneGaussian: A new hybrid representation for single-view 3D reconstruction.

LODS. Official code for ECCV 2024 paper: Learn to Optimize Denoising Scores A Unified and Improved Diffusion Prior for 3D Generation

T2I-CompBench. [Neurips 2023 & TPAMI] T2I-CompBench (++) for Compositional Text-to-image Generation Evaluation

RPG-DiffusionMaster. [ICML 2024] Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs (RPG)

360DVD. [CVPR2024] 360DVD: Controllable Panorama Video Generation with 360-Degree Video Diffusion Model

GPTEval3D. [ CVPR 2024 ] Implementation for "GPT-4V(ision) is a Human-Aligned Evaluator for Text-to-3D Generation"

3D-Gaussian-Splatting-Papers. 3D高斯论文，持续更新，欢迎交流讨论。

supersplat. 3D Gaussian Splat Editor

threestudio-3dgs. 3D Gaussian Splatting extension of threestudio.

richdreamer. [CVPR2024 (Highlight)] RichDreamer: A Generalizable Normal-Depth Diffusion Model for Detail Richness in Text-to-3D. Live Demo：https://modelscope.cn/studios/Damo_XR_Lab/3D_AIGC

OpenLRM. An open-source impl. of Large Reconstruction Models

Machine-Mindset. An MBTI Exploration of Large Language Models

repaint123. Official implementation of Repaint123: Fast and High-quality One Image to 3D Generation with Progressive Controllable 2D Repainting (ECCV 2024)

GaussianDreamer. [CVPR 2024] GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models

HumanGaussian. [CVPR 2024 Highlight] Code for "HumanGaussian: Text-Driven 3D Human Generation with Gaussian Splatting"

EditGuard. [CVPR 2024] EditGuard: Versatile Image Watermarking for Tamper Localization and Copyright Protection

VTB. Official implementation of "A Simple Visual-Textual Baseline for Pedestrian Attribute Recognition" [TCSVT 2022]

GraphDreamer. [CVPR'24] GraphDreamer: a novel framework of generating compositional 3D scenes from scene graphs.