A machine learning accelerator core designed for energy-efficient AI at the edge.

High performance self-hosted photo and video management solution.

本项目是一个通过文字生成图片的项目，基于开源模型Stable Diffusion V1.5生成可以在手机的CPU和NPU上运行的模型，包括其配套的模型运行框架。

[ICLR 2021, Spotlight] Large Scale Image Completion via Co-Modulated Generative Adversarial Networks

An unofficial bicubic++ repo

The state-of-the-art VSR

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

[ECCV 2024] PowerPaint, a versatile image inpainting model that supports text-guided object inpainting, object removal, image outpainting and shape-guided object inpainting with only a single model…

Official repository for MixFaceNets: Extremely Efficient Face Recognition Networks

Pico TTS: text to speech voice sinthesizer from SVox, included in Android AOSP

Code and models for the paper "One Transformer Fits All Distributions in Multi-Modal Diffusion"

NSNet2 Deep Noise Suppression (DNS) package

TinyML AI inference library

Reference implementations of MLPerf® inference benchmarks

This is the open-source version of TinyTS. The code is dirty so far. We may clean the code in the future.

A unified library for object tracking featuring clean room re-implementations of leading multi-object tracking algorithms

Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.

ONNXim is a fast cycle-level simulator that can model multi-core NPUs for DNN inference

Implementation of Imagen, Google's Text-to-Image Neural Network, in Pytorch

Slides with modifications for a course at Tsinghua University.

High-Resolution Image Synthesis with Latent Diffusion Models

The project can achieve FCWS, LDWS, and LKAS functions solely using only visual sensors. using YOLOv5 / YOLOv5-lite / YOLOv6 / YOLOv7 / YOLOv8 / YOLOv9 / EfficientDet and Ultra-Fast-Lane-Detection-…

To train deep convolutional neural networks, the input data and the activations need to be kept in memory. Given the limited memory available in current GPUs, this limits the maximum dimensions of …

Use streaming to train whole-slides images with single image-level labels, by reducing GPU memory requirements with 99%.

(CVPR 2025 Highlight) Official repository of paper "AODRaw: Towards RAW Object Detection in Diverse Conditions" (https://arxiv.org/pdf/2411.15678)

LLM-Powered GUI Agents in Phone Automation: Surveying Progress and Prospects

UFPR-ALPR: a dataset for license plate detection and recognition that includes 4,500 fully annotated images acquired in real-world scenarios where both the vehicle and the camera (inside another ve…