megvii-research

The state-of-the-art image restoration model without nonlinear activation functions.

PyTorch implementation of Multi-Label Image Recognition with Graph Convolutional Networks, CVPR 2019.

[ECCV2022] PETR: Position Embedding Transformation for Multi-View 3D Object Detection & [ICCV2023] PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images

The official implementation of [CVPR2022] Decoupled Knowledge Distillation https://arxiv.org/abs/2203.08679 and [ICCV2023] DOT: A Distillation-Oriented Trainer...

A series of basic algorithms that are useful for video understanding, including Single Object Tracking (SOT), Video Object Segmentation (VOS) and so on.

[ECCV 2024] HiDiffusion: Increases the resolution and speed of your diffusion model by only adding a single line of code!

[ECCV2022] MOTR: End-to-End Multiple-Object Tracking with TRansformer

IJCAI2023 - Collaborative Neural Rendering using Anime Character Sheets

The official PyTorch implementation of paper BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognition

Official MegEngine implementation of CREStereo(CVPR 2022 Oral).

Paint artistic patterns using random neural network.

[CVPR2023] MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectors

Homepage for the joint course of Megvii Inc. and Peking University on Deep Learning.

[IJCAI 2022] FQ-ViT: Post-Training Quantization for Fully Quantized Vision Transformer

An official implementation of the Anchor DETR.

Multi-Stage Pose Network

A model compression and acceleration toolbox based on pytorch.

Maybe the first academic open work on stereo 3D SSC method with vision-only input.

(CVPR 2022) TransMVSNet: Global Context-aware Multi-view Stereo Network with Transformers.

Official Code of Paper "Reversible Column Networks" "RevColv2"

Test-time Local Converter

[IROS2025]This is the offical implementation of the paper "MCTrack: A Unified 3D Multi-Object Tracking Framework for Autonomous Driving"

Official PyTorch implementation of the paper "Deep Constrained Least Squares for Blind Image Super-Resolution", CVPR 2022.

[AAAI2024] Far3D: Expanding the Horizon for Surround-view 3D Object Detection

"SOLQ: Segmenting Objects by Learning Queries", SOLQ is an end-to-end instance segmentation framework with Transformer.

Course materials for Introduction to Neural 3D Reconstruction

[TPAMI 2023 / ACMMM 2022 Best Paper Runner-Up Award] Learnability Enhancement for Low-light Raw Denoising: Where Paired Real Data Meets Noise Modeling (a Data Perspective)

Official implementation of ID-unaware Deepfake Detection Model

Tutorials for deep learning course here:

[IJCV] Official Implementation of PVD and PVDAL: http://sk-fun.fun/PVD-AL/

Megvii FILE Library - Working with Files in Python same as the standard library

[CVPR 2020] DPGN: Distribution Propagation Graph Network for Few-shot Learning.

CVPR2023 (highlight) - UniDistill: A Universal Cross-Modality Knowledge Distillation Framework for 3D Object Detection in Bird's-Eye View

The official PyTorch implementation of paper Exploring Categorical Regularization for Domain Adaptive Object Detection (CR-DA-DET)

[CVPR2022] Tree Energy Loss: Towards Sparsely Annotated Semantic Segmentation

The official implementation of the ECCV 2022 Oral paper: RealFlow: EM-based Realistic Optical Flow Dataset Generation from Videos

The official MegEngine implementation of the ECCV 2022 paper: Ghost-free High Dynamic Range Imaging with Context-aware Transformer

A hyperparameter manager for deep learning experiments.

Official implementation of the paper "Progressive End-to-End Object Detection in Crowded Scenes"

torch version of instant-ngp, image rendering

[SafeAI'21] Feature Space Singularity for Out-of-Distribution Detection.

PyTorch implementation of SSQL (Accepted to ECCV2022 oral presentation)

The official PyTorch implementation of the paper "Motion Basis Learning for Unsupervised Deep Homography Estimation with Subspace Projection".

Slides with modifications for a course at Tsinghua University.

📍 Official repository of paper "ProtoCLIP: Prototypical Contrastive Language Image Pretraining" (IEEE TNNLS 2023)

This is the official implementation of HomoGAN, CVPR2022

Official Implementation of the detection self-distillation framework LGD.