Solve puzzles. Learn CUDA.

FastAPI RabbitMQ Dramatiq Demo

official implementation for the paper 'Representation Learning and Identity Adversarial Training for Facial Behavior Understanding'

Convert images of LaTex math equations into LaTex code.

MixTeX multimodal LaTeX, ZhEn, and, Table OCR. It performs efficient CPU-based inference in a local offline on Windows.

🌊 [ECCV'24 Oral] MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images

Parallelized triangle mesh --> continuous signed distance field on CPU

End-to-end face detection, cropping, norm estimation, and landmark detection in a single onnx model

RTMPose series (RTMPose, DWPose, RTMO, RTMW) without mmcv, mmpose, mmdet etc.

SAM2-UNet: Segment Anything 2 Makes Strong Encoder for Natural and Medical Image Segmentation

[CAAI AIR'24] Bilateral Reference for High-Resolution Dichotomous Image Segmentation

High-resolution models for human tasks.

FaRL for Facial Representation Learning [Official, CVPR 2022]

CatVTON is a simple and efficient virtual try-on diffusion model with 1) Lightweight Network (899.06M parameters totally), 2) Parameter-Efficient Training (49.57M parameters trainable) and 3) Simpl…

Stable-Hair: Real-World Hair Transfer via Diffusion Model

Goliath Dataset and Official PyTorch Implementation of RelightableHands, Relightable Gaussian Codec Avatars, and Driving-Signal Aware Full-Body Avatars.

#1 Locally hosted web application that allows you to perform various operations on PDF files

👔IMAGDressing👔: Interactive Modular Apparel Generation for Virtual Dressing

🛡️ Open-source and next-generation Web Application Firewall (WAF)

[NeurIPS 2024] Code release for "Segment Anything without Supervision"

SAiD: Blendshape-based Audio-Driven Speech Animation with Diffusion

Bring portraits to life!

Official implementation of "Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling"

Open source real-time translation app for Android that runs locally

CUDA accelerated rasterization of gaussian splatting

Mesh generation by the adaptive process

68/21 Landmark points for Basel Face Model (3DMM)

Synthetic Faces High Quality (SFHQ) Dataset. 425,000 curated 1024x1024 synthetic face images