Renrui Zhang

PhD in MMLab, CUHK, Peking University.

GitHub 17 repos 805 followers

Explained projects

Point-M2AE: The 3D Masked Autoencoder That Thinks in Pyramids

A deep dive into how hierarchical masking, token merging, and custom CUDA kernels let a point-cloud model learn shape efficiently without flattening geometry into a plain transformer.

9 min read · Jun 17, 2026

Yet to be explained

Personalize-SAM

Personalize Segment Anything Model (SAM) with 1 shot in 10 seconds

Python1.7k stars

Explain

Point-NN

[CVPR 2023] Parameter is Not All You Need: Starting from Non-Parametric Networks for 3D Point Cloud Analysis

Python539 stars

Explain

MonoDETR

[ICCV 2023] The first DETR model for monocular 3D object detection with depth-guided transformer

Python447 stars

Explain

PointCLIP

[CVPR 2022] PointCLIP: Point Cloud Understanding by CLIP

Python412 stars

Explain

I2P-MAE

[CVPR 2023] Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders

Python230 stars

Explain

MathVerse

[ECCV 2024] Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?

Python183 stars

Explain

MAVIS

[ICLR 2025] Mathematical Visual Instruction Tuning for Multi-modal Large Language Models

157 stars

Explain

LLaMA-Adapter

Fine-tuning LLaMA to follow Instructions within 1 Hour and 1.2M Parameters

92 stars

Explain

CaFo

[CVPR 2023] Prompt, Generate, then Cache: Cascade of Foundation Models makes Strong Few-shot Learners

45 stars

Explain

MonoDETR-MV

The multi-view version of MonoDETR on nuScenes dataset

21 stars

Explain

ZrrSkywalker

5 stars

Explain

CALIP

Enhancing Zero-shot CLIP with Cross-Modality Attention

4 stars

Explain

ZrrSkywalker.github.io

HTML1 stars

Explain