Zhenyu Li

edm2. EDM2 and Autoguidance -- Official PyTorch implementation

DivRL. Official implementation of paper DivRL: Disentangled Self-Similarity Rewards for Diverse Subject-Driven Generation (ECCV 2026)

Internal-Guidance. CVPR 2026 (Highlight)-Guiding a Diffusion Transformer with the Internal Dynamics of Itself (IG)

Awesome-Video-World-Models-with-AR-Diffusion. A Curated List of Awesome Video World Models with AR Diffusion: Covering Algorithms, Applications, and Infrastructure, Aimed at Serving as a Comprehensive Resource for Researchers, Practitioners, and Enthusiasts.

PartEdit. Jupyter Notebook

NearID. Official Implementation for NearID: Identity Representation Learning via Near-identity Distractors

black. The uncompromising Python code formatter

NeoVerse. [CVPR 2026 Highlight & Best Paper of VideoWorldModel Workshop] NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videos

V2-SAM. [CVPR'26 Highlight] Official Code for “V²-SAM: Marrying SAM2 with Multi-Prompt Experts for Cross-View Object Correspondence”

EgoSound. Official release of 'EgoSound: Benchmarking Sound Understanding in Egocentric Videos[CVPR 2026 Highlight]'

EgoNight. EgoNight is the first comprehensive benchmark designed to evaluate egocentric vision understanding in low-light and nighttime conditions—a critical gap in current research.

codegen-metrics. Replication package for evaluation of code generation metrics

Any-Resolution-Any-Geometry. Python

ReNoV.

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

PoseGAM. A novel multi-view feedforward network that enables direct and robust object pose estimation from a query image.

Room_Envelopes. Room Envelopes layout dataset generation code

Depth-Anything-3. Depth Anything 3

cmrw. Official PyTorch implementation of Self-Supervised Spatial Correspondence Across Modalities, CVPR 2025.

co-tracker. CoTracker is a model for tracking any point (pixel) on a video.

True-Story-of-Pangu. 诺亚盘古大模型研发背后的真正的心酸与黑暗的故事。

content_aware_tiles. A flexible training-free diffusion-based method for generating tileable image sets, including self-tiling images, stochastic self-tiling images, and Wang tile textures.

nanoVLM. The simplest, fastest repository for training/finetuning small-sized VLMs.

zotero-arxiv-daily. Recommend new arxiv papers of your interest daily according to your Zotero libarary.

mpsfm. MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion (CVPR 2025)

2d-gaussian-splatting. [SIGGRAPH'24] 2D Gaussian Splatting for Geometrically Accurate Radiance Fields

GenFusion. [CVPR 2025] GenFusion: Closing the Loop between Reconstruction and Generation via Videos

Feat2GS. [CVPR2025] Feat2GS: Probing Visual Foundation Models with Gaussian Splatting

LaRa. [ECCV 2024] Efficient Large-Baseline Radiance Fields, a feed-forward 2DGS model

Gen3DSR. Gen3DSR: Generalizable 3D Scene Reconstruction via Divide and Conquer from a Single View, 3DV2025

placeit3d. [ICCV 2025] PlaceIt3D: Language-Guided Object Placement in Real 3D Scenes

DeDoDe. [3DV 2024 Oral] DeDoDe 🎶 Detect, Don't Describe --- Describe, Don't Detect, for Local Feature Matching

KEEP. [ECCV'24] Kalman-Inspired Feature Propagation for Video Face Super-Resolution

SPIdepth. [CVPR 2025] Strengthened Pose Information for self-supervised monocular depth estimation. SPIdepth refines the pose network to improve depth prediction accuracy, achieving state-of-the-art results on benchmarks like KITTI, Cityscapes, and Make3D.

lari. [ICML 2026] 🎨 Occluded 3D Scene Reconstruction from a Single Image.

7-Scenes-Calibration. Calibration RGB / Depth for the 7-Scenes dataset

unidet3d. [AAAI2025] UniDet3D: Multi-dataset Indoor 3D Object Detection

UniMODE. Python

One-Look-is-Enough. [ICCV 2025] Official repository of "One Look is Enough"

LlamaFactory. Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

EditCLIP. Implementation of paper EditCLIP: Representation Learning for Image Editing (ICCV 2025)

SpatialLM. [NeurIPS 2025] SpatialLM: Training Large Language Models for Structured Indoor Modeling

CUT3R. Official implementation of Continuous 3D Perception Model with Persistent State

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

PseDet. [ICLR 2025] PseDet: Revisiting the Power of Pseudo Label in Incremental Object Detection

UniQA-3D. [3DV25] Official code for "Towards Foundation Models for 3D Vision: How Close Are We?"

RDED. [CVPR 2024] On the Diversity and Realism of Distilled Dataset: An Efficient Dataset Distillation Paradigm

IGEV. [TPAMI 2025 & CVPR 2023] Iterative Geometry Encoding Volume for Stereo Matching

tread. Python

DoubleDiffusion_3D_Mesh. Diffusion generation on Mesh toolbox

awesome-vlm-architectures. Famous Vision Language Models and Their Architectures

PatchRefinerV2. [ICLR 2026] PatchRefiner V2: Fast and Lightweight Real-Domain High-Resolution Metric Depth Estimation

arXiv_recbot. A Telegram bot to recommend arXiv papers

Amodal-Depth-Anything. [ICCV 2025] Amodal Depth Anything: Amodal Depth Estimation in the Wild

Awesome-LLM-3D. Awesome-LLM-3D: a curated list of Multi-modal Large Language Model in 3D world Resources

diffindscene. [CVPR2024] DiffInDScene: Diffusion-based High-Quality 3D Indoor Scene Generation

flux. Official inference repo for FLUX.1 models

CVinW_Readings. A collection of papers on the topic of ``Computer Vision in the Wild (CVinW)''

awesome-dust3r. 🌟A curated list of DUSt3R-related papers and resources, tracking recent advancements using this geometric foundation model.

depthsplat. [CVPR'25] DepthSplat: Connecting Gaussian Splatting and Depth

ViewFormer-Occ. [ECCV 2024] ViewFormer: Exploring Spatiotemporal Modeling for Multi-View 3D Occupancy Perception via View-Guided Transformers

DIA. Dissolving Is Amplifying: Towards Fine-Grained Anomaly Detection https://arxiv.org/pdf/2302.14696

ImmersePro. [ICML2026] End-to-End Stereo Video Synthesis Via Implicit Disparity Learning

wild-gaussian-splatting. DUSt3R + Gaussian Splatting

ml-depth-pro. Depth Pro: Sharp Monocular Metric Depth in Less Than a Second.

FreeReg. [ICLR 2024] FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimators

OPUS. OPUS: Occupancy Prediction Using a Sparse Set

spann3r. [3DV'25 Award Candidate] 3D Reconstruction with Spatial Memory

splatt3r. Official repository for Splatt3R: Zero-shot Gaussian Splatting from Uncalibrated Image Pairs

DiffusionDepth. PyTorch Implementation of introducing diffusion approach to 3D depth perception ECCV 2024

ECCV-2024-Papers-Autonomous-Driving. ECCV 2024 Paper List about Autonomous Driving

BadMerging. [CCS 2024] "BadMerging: Backdoor Attacks Against Model Merging": official code implementation.

HybridDepth. Official implementation for HybridDepth Model [WACV 2025, ISMAR 2024]

flash3d. Flash3D: Feed-Forward Generalisable 3D Scene Reconstruction from a Single Image

MindSearch. 🔍 An LLM-based Multi-agent Framework of Web Search Engine (like Perplexity.ai Pro and SearchGPT)

sam2. The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

MindSearch_tmp. LLM-based Multi-agent Framework of AI Search Engine

Diffusion4RobustDepth. [ECCV 2024] Diffusion Models for Monocular Depth Estimation: Overcoming Challenging Conditions

UVTR. Unifying Voxel-based Representation with Transformer for 3D Object Detection (NeurIPS 2022)

Lite-Mono. [CVPR2023] Lite-Mono: A Lightweight CNN and Transformer Architecture for Self-Supervised Monocular Depth Estimation

Mono-ViFI. [ECCV 2024] Mono-ViFI: A Unified Learning Framework for Self-supervised Single- and Multi-frame Monocular Depth Estimation

Diffusion-TTA. Diffusion-TTA improves pre-trained discriminative models such as image classifiers or segmentors using pre-trained generative models.

GeoBench. A toolbox for benchmarking SOTA discriminative and generative geometry estimation models.

Summer2026-Internships. Summer 2026 software engineering, data science, AI, quant, product management, and hardware internship postings. Updated daily by Simplify and Pitt CSC.

StabilityIndex. [ECCV 2024] Towards Stable 3D Object Detection

sap3d. [CVPR 2024 Hightlight] Code release for "The More You See in 2D, the More You Perceive in 3D"

bidastereo. [ECCV 2024] Match-Stereo-Videos: Bidirectional Alignment for Consistent Dynamic Stereo Matching.

PatchRefiner. [ECCV 2024] Leveraging Synthetic Data for Real-Domain High-Resolution Monocular Metric Depth Estimation

TripoSR. TripoSR: Fast 3D Object Reconstruction from a Single Image

RayDiffusion. Code for "Cameras as Rays"

pix2gestalt. Code for the paper "pix2gestalt: Amodal Segmentation by Synthesizing Wholes" (CVPR 2024)

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

VAR. [NeurIPS 2024 Best Paper Award][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ultra-simple, user-friendly yet state-of-the-art* codebase for autoregressive image generation!

VidSeg_diffusion. Implementation of Zero-Shot Video Semantic Segmentation [CVPR 2025]

depth-fm. [AAAI 2025, Oral] DepthFM: Fast Monocular Depth Estimation with Flow Matching

TemporallyConsistentDepth. Code for our CVPR 2023 paper on online, temporally consistent depth estimation.

Entity. EntitySeg Toolbox: Towards Open-World and High-Quality Image Segmentation

ShareGPT4Video. [NeurIPS 2024] An official implementation of "ShareGPT4Video: Improving Video Understanding and Generation with Better Captions"

edm. Elucidating the Design Space of Diffusion-Based Generative Models (EDM)

HowToCook. Programmer's guide about how to cook at home.

OccSora. OccSora: 4D Occupancy Generation Models as World Simulators for Autonomous Driving

EcoDepth. [CVPR'2024] Official implementation of the paper "ECoDepth: Effective Conditioning of Diffusion Models for Monocular Depth Estimation"

dtd. Dusk Till Dawn: Self-supervised Nighttime Stereo Depth Estimation using Visual Foundation Models

EVP. [ECCV 2024] EVP model for metric depth estimation from a single image and referring segmentation

GenPercept. [ICLR2025] GenPercept: Diffusion Models Trained with Large Data Are Transferable Visual Models

segment-anything. The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

flowmap. [3DV 2025] Code for "FlowMap: High-Quality Camera Poses, Intrinsics, and Depth via Gradient Descent" by Cameron Smith*, David Charatan*, Ayush Tewari, and Vincent Sitzmann

probe3d. [CVPR 2024] Probing the 3D Awareness of Visual Foundation Models

mickey. [CVPR 2024 - Oral] Matching 2D Images in 3D: Metric Relative Pose from Metric Correspondences

infinigen. Infinite Photorealistic Worlds using Procedural Generation

chatgpt-prompts-for-academic-writing. This list of writing prompts covers a range of topics and tasks, including brainstorming research ideas, improving language and style, conducting literature reviews, and developing research plans.

matterport_utils. Utilities to prepare Matterport panoramic data for semantic and instance segmentation

PlainMamba. [BMVC 2024] PlainMamba: Improving Non-hierarchical Mamba in Visual Recognition

slowtv_monodepth. Official repository for the ICCV2023 paper "Kick Back & Relax: Learning to Reconstruct the World by Watching SlowTV"

GeoWizard. [ECCV'24] GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Image

controlnet_aux. Python

T-Eval. T-Eval: Evaluating Tool Utilization Capability of Large Language Models Step by Step

DCNv4. [CVPR 2024] Deformable Convolution v4

mamba. Mamba SSM architecture

prisma. Computational photography pipeline that performs multiple inferences from any image or video.

SegRefiner. SegRefiner: Towards Model-Agnostic Segmentation Refinement with Discrete Diffusion Process

Depth-Anything. [CVPR 2024] Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data. Foundation Model for Monocular Depth Estimation

RGBD_Semantic_Segmentation_PyTorch. [ECCV 2020] PyTorch Implementation of some RGBD Semantic Segmentation models.

T-Eval. [ACL2024] T-Eval: Evaluating Tool Utilization Capability of Large Language Models Step by Step

stable-diffusion-webui. Stable Diffusion web UI

FlashMHA. An simple pytorch implementation of Flash MultiHead Attention

GW2Radial. A customizable radial menu overlay addon for Guild Wars 2.

GW2-Addon-Manager. Installer and updater for Guild Wars 2 addons.

FrozenRecon. [ICCV2023] 🧊FrozenRecon: Pose-free 3D Scene Reconstruction with Frozen Depth Models

Amphion. Amphion (/æmˈfaɪən/) is a toolkit for Audio, Music, and Speech Generation. Its purpose is to support reproducible research and help junior researchers and engineers get started in the field of audio, music, and speech generation research and development.

vppdc. [3DV 2024] Revisiting Depth Completion from a Stereo Matching Perspective for Cross-domain Generalization

VoxelKP. [ICCV 2025] A Voxel-based Network Architecture for Human Keypoint Estimation in LiDAR Data

Monocular-Depth-Estimation-Toolbox. Monocular Depth Estimation Toolbox based on MMSegmentation.

Marigold. [CVPR 2024 - Oral, Best Paper Award Candidate] Marigold: Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation

LooseControl. Lifting ControlNet for Generalized Depth Conditioning

PatchFusion. [CVPR 2024] An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation

EDGE. Official PyTorch Implementation of EDGE (CVPR 2023)

pytorch-template. PyTorch deep learning projects made easy.

hydra. Hydra is a framework for elegantly configuring complex applications

lightning-hydra-template. PyTorch Lightning + Hydra. A very user-friendly template for ML experimentation. ⚡🔥⚡

pytorch-lightning-template. An easy/swift-to-adapt PyTorch-Lighting template. 套壳模板，简单易用，稍改原来Pytorch代码，即可适配Lightning。You can translate your previous Pytorch code much easier using this template, and keep your freedom to edit all the functions as well. Big-project-friendly as well. No need to rewrite your config in hydra.

4d-occ-forecasting. CVPR 2023: Official code for `Point Cloud Forecasting as a Proxy for 4D Occupancy Forecasting'

ml-interviews-book. https://huyenchip.com/ml-interviews-book/

hierarchical_perceiver. Python

vidar. Python

MPI-Flow. Official Code of "[ICCV 2023] MPI-Flow: Learning Realistic Optical Flow with Multiplane Images"

DETRDistill. [ICCV2023] DETRDistill: A Universal Knowledge Distillation Framework for DETR-families

Startup-CTO-Handbook. The Startup CTO's Handbook, a book covering leadership, management and technical topics for leaders of software engineering teams

DynamicRouting. Learning Dynamic Routing for Semantic Segmentation

ZoeDepth. Metric depth estimation from a single image

camel. 🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org

mmsegmentation. OpenMMLab Semantic Segmentation Toolbox and Benchmark.

instant-ngp. Instant neural graphics primitives: lightning fast NeRF and more

learning_research. 本人的科研经验

idisc. iDisc: Internal Discretization for Monocular Depth Estimation [CVPR 2023]

depthgan. [ECCV 2022 Oral] 3D-Aware Indoor Scene Synthesis with Depth Priors

ImageBind. ImageBind One Embedding Space to Bind Them All

mmagic. OpenMMLab Multimodal Advanced, Generative, and Intelligent Creation Toolbox. Unlock the magic 🪄: Generative-AI (AIGC), easy-to-use APIs, awsome model zoo, diffusion models, for text-to-image generation, image/video restoration/enhancement, etc.

HieraSeg. CVPR2022 - Deep Hierarchical Semantic Segmentation - A structured, pixel-wise description of visual scenes in terms of the class hierarchy.

Lift3D. CVPR 2023 Use NeRF-generated images to train your model.

annotated_deep_learning_paper_implementations. 🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

PointCert. [CVPR 2023] "PointCert: Point Cloud Classification with Deterministic Certified Robustness Guarantees": official code implementation.

occupancy-for-nuscenes. 3D occupancy

Satellite-Communication-system-project. The project here contains the baseband simulation the satellite communication system with DVB-S standards.

CVPR2023-3D-Occupancy-Prediction. CVPR2023-Occupancy-Prediction-Challenge

HDMapNet. Python

BEVDistill. [ICLR 2023] BEVDistill: Cross-Modal BEV Distillation for Multi-View 3D Object Detection

brace. BRACE: The Breakdancing Competition Dataset for Dance Motion Synthesis

MagFace. MagFace: A Universal Representation for Face Recognition and Quality Assessment, CVPR2021, Oral

PixelFormer. Python

DEVIANT. [ECCV 2022] Official PyTorch Code of DEVIANT: Depth Equivariant Network for Monocular 3D Object Detection

hyperparameter-optimization. Implementation of Bayesian Hyperparameter Optimization of Machine Learning Algorithms

EPro-PnP. [CVPR 2022 Best Student Paper] EPro-PnP: Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimation

AutoAlignV2. [ECCV2022, IJCAI2022] AutoAlignV2: Deformable Feature Aggregation for Dynamic Multi-Modal 3D Object Detection

MaGNet. [CVPR 2022 Oral] Multi-View Depth Estimation by Fusing Single-View Depth Probability with Multi-View Geometry

FillDepth. The code of 'Towards Domain-agnostic depth completion'

pifuhd. High-Resolution 3D Human Digitization from A Single Image.

PIFu. This repository contains the code for the paper "PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization"

Depth-from-Motion. [ECCV 2022 oral] Monocular 3D Object Detection with Depth from Motion

robust_loss_pytorch. A pytorch port of google-research/google-research/robust_loss/

Graph-DETR4D. [ACM MM2022, TIP2024] Graph-DETR Series for Multi-View 3D Object Detection

SoftTeacher. Semi-Supervised Learning, Object Detection, ICCV2021

awesome_3DReconstruction_list. A curated list of papers & resources linked to 3D reconstruction from images.

Automold--Road-Augmentation-Library. This library augments road images to introduce various real world scenarios that pose challenges for training neural networks of Autonomous vehicles. Automold is created to train CNNs in specific weather and road conditions.

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

ctrl-uda. An implementation of our work "Learning to Relate Depth and Semantics for Unsupervised Domain Adaptation".

S2R-DepthNet. Python

awesome-neural-rendering. Resources of Neural Rendering

mmrazor. OpenMMLab Model Compression Toolbox and Benchmark.

CWD. Channel-wise Distillation for Semantic Segmentation

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

MUTR3D. Code for Paper, MUTR3D: A Multi-camera Tracking Framework via 3D-to-2D Queries. https://tsinghua-mars-lab.github.io/mutr3d/

AB3DMOT. (IROS 2020, ECCVW 2020) Official Python Implementation for "3D Multi-Object Tracking: A Baseline and New Evaluation Metrics"

BEVDet. Code base of the BEVDet series .

BoostingDepth. Official implementation of "Towards 3D Scene Reconstruction from Locally Scale-Aligned Monocular Video Depth (Boosting Monocular Depth Estimation with Sparse Guided Points)"

awesome-multiple-object-tracking. Resources for Multiple Object Tracking (MOT)

OpenPCDet. OpenPCDet Toolbox for LiDAR-based 3D Object Detection.

CaDDN. Categorical Depth Distribution Network for Monocular 3D Object Detection (CVPR 2021 Oral)

qd-3dt. Official implementation of Monocular Quasi-Dense 3D Object Tracking, TPAMI 2022

Paper-Writing-Tips. MLNLP社区用来帮助大家避免论文投稿小错误的整理仓库。 Paper Writing Tips