Official Implementation of VideoDPO

A unified LiDAR-visual system achieving geometrically consistent photorealistic rendering and high-granularity surface reconstruction.

Official PyTorch implementation of paper “InsViE-1M: Effective Instruction-based Video Editing with Elaborate Dataset Construction”

[CVPR 2025] TexGaussian: Generating High-quality PBR Material via Octree-based 3D Gaussian Splatting

[CVPR 2025] PanoGS: Gaussian-based Panoptic Segmentation for 3D Open Vocabulary Scene Understanding

[CVPR-2025] The official code of HunyuanPortrait: Implicit Condition Control for Enhanced Portrait Animation

Official Implementation of Video-T1: Test-Time Scaling for Video Generation

OpenVLThinker: An Early Exploration to Vision-Language Reasoning via Iterative Self-Improvement

[CVPR 2025] UniK3D: Universal Camera Monocular 3D Estimation

This is the official implementation of Video Generation part of This&That: Language-Gesture Controlled Video Generation for Robot Planning (ICRA 2025)

[ICLR 2025] The official implementation of SSR

[ECCV2024] Official Implementation of "NVS-Adapter: Plug-and-Play Novel View Synthesis from a Single Image"

Code for: "Long-Context Autoregressive Video Modeling with Next-Frame Prediction"

Official implementation of DreamRenderer: Taming Multi-Instance Attribute Control in Large-Scale Text-to-Image Models

[CVPR 2025] Diffusion-4K: Ultra-High-Resolution Image Synthesis with Latent Diffusion Models

PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slides

A curated list of image inpainting and video inpainting papers and resources

Blender toolbox for rendering 3D scenes.

Vulkan-based Gaussian Splatting viewer, and python binding

This is the code of the CVPR 2025 paper "Hardware-Rasterized Ray-Based Gaussian Splatting"

[ICLR 2025] IterComp: Iterative Composition-Aware Feedback Learning from Model Gallery for Text-to-Image Generation

An Efficient Text-to-Image Generation Pretrain Pipeline

Official implementation of 🛸 "UFO: A Unified Approach to Fine-grained Visual Perception via Open-ended Language Interface"

[CVPR 2025] Zero-1-to-A: Zero-Shot One Image to Animatable Head Avatars Using Video Diffusion

OffsetOPT: Explicit Surface Reconstruction without Normals (CVPR2025)

[CVPR 2025] Official code of "From Zero to Detail: Deconstructing Ultra-High-Definition Image Restoration from Progressive Spectral Perspective"

[CVPR 2025] Multi-focal Conditioned Latent Diffusion for Person Image Synthesis

Cosmos-Predict1 is a collection of general-purpose world foundation models for Physical AI that can be fine-tuned into customized world models for downstream applications.